[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"summary-dd59ec1e4e966fad-gemma-4-26b-a4b-4b-active-moe-for-multimodal-ai-summary":3,"summaries-facets-categories":102,"summary-related-dd59ec1e4e966fad-gemma-4-26b-a4b-4b-active-moe-for-multimodal-ai-summary":3671},{"id":4,"title":5,"ai":6,"body":13,"categories":54,"created_at":56,"date_modified":56,"description":47,"extension":57,"faq":56,"featured":58,"kicker_label":56,"meta":59,"navigation":84,"path":85,"published_at":56,"question":56,"scraped_at":86,"seo":87,"sitemap":88,"source_id":89,"source_name":90,"source_type":91,"source_url":92,"stem":93,"tags":94,"thumbnail_url":56,"tldr":99,"tweet":56,"unknown_tags":100,"__hash__":101},"summaries\u002Fsummaries\u002Fdd59ec1e4e966fad-gemma-4-26b-a4b-4b-active-moe-for-multimodal-ai-summary.md","Gemma 4 26B A4B: 4B Active MoE for Multimodal AI",{"provider":7,"model":8,"input_tokens":9,"output_tokens":10,"processing_time_ms":11,"cost_usd":12},"openrouter","x-ai\u002Fgrok-4.1-fast",7989,2096,22810,0.0026233,{"type":14,"value":15,"toc":46},"minimark",[16,21,25,29,32,36,39,43],[17,18,20],"h2",{"id":19},"moe-architecture-delivers-dense-like-speed-at-scale","MoE Architecture Delivers Dense-Like Speed at Scale",[22,23,24],"p",{},"Gemma 4 26B A4B-it employs a Mixture-of-Experts (MoE) design with 25.2B total parameters but only 3.8B active per inference (8 active \u002F 128 total experts + 1 shared), running as fast as a 4B dense model while outperforming Gemma 3 27B. It features 30 layers, 1024-token sliding window, 256K context via hybrid attention (local sliding + global with unified KV and p-RoPE), and 262K vocab. Supports text\u002Fimage inputs (~550M vision params); audio\u002Fvideo on smaller E2B\u002FE4B. Dense siblings: E2B (2.3B effective\u002F5.1B total, 35 layers, 128K context, text\u002Fimage\u002Faudio), E4B (4.5B effective\u002F8B total, 42 layers, same context\u002Fmodalities), 31B (30.7B params, 60 layers, 256K text\u002Fimage). Per-Layer Embeddings (PLE) on small models boost efficiency for on-device use. Native system prompts, function-calling, and configurable thinking modes (\u003C|think|>, \u003C|channel|thought\n\u003Cchannel|>) enable agentic workflows.",[17,26,28],{"id":27},"benchmark-leadership-in-reasoning-coding-multimodality","Benchmark Leadership in Reasoning, Coding, Multimodality",[22,30,31],{},"Instruction-tuned models crush benchmarks: 26B A4B-it scores 82.6% MMLU Pro, 88.3% AIME 2026 (no tools), 77.1% LiveCodeBench v6, 1718 Codeforces ELO, 82.3% GPQA Diamond, 86.3% MMMLU, 73.8% MMMU Pro, 82.4% MATH-Vision; long-context MRCR v2 128K at 44.1%. Larger 31B hits 85.2% MMLU Pro, 89.2% AIME, 80.0% LiveCodeBench, 2150 ELO. Small E4B\u002FE2B: 69.4%\u002F60.0% MMLU Pro, audio CoVoST 35.54%\u002F33.47%. All beat Gemma 3 27B (no think) by wide margins, e.g., Tau2 68.2% vs 16.2%, BigBench Hard 64.8% vs 19.3%. Vision excels in OmniDocBench (0.149 avg edit distance, lower better). Use thinking mode (enable_thinking=True) for complex reasoning; parse with processor.parse_response.",[17,33,35],{"id":34},"transformers-integration-for-text-image-audio-video","Transformers Integration for Text, Image, Audio, Video",[22,37,38],{},"Load via pip install -U transformers torch accelerate; use AutoProcessor\u002FAutoModelForCausalLM (text) or AutoModelForMultimodalLM (multimodal). Example text gen: apply_chat_template on messages with system\u002Fuser roles, generate max_new_tokens=1024, decode\u002Fparse. Multimodal: add {\"type\": \"image\u002Faudio\u002Fvideo\", \"url\": \"path\"} before text in user content; audio needs librosa, video torchcodec\u002Ftorchvision. Sampling: temperature=1.0, top_p=0.95, top_k=64. Audio max 30s, video 60s (1fps). Prompts: image variable resolution\u002Faspect via token budget; audio transcription e.g., \"Transcribe in {LANG}, digits only, no newlines.\" Multi-turn via standard roles. Modality order: non-text before text.",[17,40,42],{"id":41},"safety-gains-with-provenance-focus","Safety Gains with Provenance Focus",[22,44,45],{},"Trained on web\u002Fcode\u002Fimages\u002Faudio (cutoff Jan 2025) with dedup, filtering, PII removal. Matches Gemini safety evals: minimal violations (text\u002Fimage), low unjustified refusals vs Gemma 3. Aligns Google AI Principles; risks like bias\u002Fhallucinations mitigated via evals\u002Fpartnerships. Intended for reasoning\u002Fcoding\u002Fagents\u002Fmultimodal apps; limits: no native video\u002Faudio on large models, potential ethical VLMs issues (fairness, misuse). Apache 2.0 licensed for enterprise\u002Fon-device deployment.",{"title":47,"searchDepth":48,"depth":48,"links":49},"",2,[50,51,52,53],{"id":19,"depth":48,"text":20},{"id":27,"depth":48,"text":28},{"id":34,"depth":48,"text":35},{"id":41,"depth":48,"text":42},[55],"AI & LLMs",null,"md",false,{"content_references":60,"triage":79},[61,66,69,72,76],{"type":62,"title":63,"url":64,"context":65},"other","Gemma 4 Launch Blog","https:\u002F\u002Fblog.google\u002Finnovation-and-ai\u002Ftechnology\u002Fdevelopers-tools\u002Fgemma-4\u002F","mentioned",{"type":62,"title":67,"url":68,"context":65},"Gemma Documentation","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore",{"type":62,"title":70,"url":71,"context":65},"Gemma 4 License","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fgemma_4_license",{"type":62,"title":73,"url":74,"context":75},"Google’s AI Principles","https:\u002F\u002Fai.google\u002Fprinciples\u002F","cited",{"type":62,"title":77,"url":78,"context":65},"Google Gemma GitHub","https:\u002F\u002Fgithub.com\u002Fgoogle-gemma",{"relevance":80,"novelty":81,"quality":81,"actionability":81,"composite":82,"reasoning":83},5,4,4.35,"Category: AI & LLMs. The article provides in-depth technical details about the Gemma 4 26B A4B model, including its architecture and performance benchmarks, which are crucial for developers looking to implement AI features. It also includes practical integration instructions, making it actionable for the audience.",true,"\u002Fsummaries\u002Fdd59ec1e4e966fad-gemma-4-26b-a4b-4b-active-moe-for-multimodal-ai-summary","2026-04-16 03:04:54",{"title":5,"description":47},{"loc":85},"dd59ec1e4e966fad","__oneoff__","article","https:\u002F\u002Fhuggingface.co\u002Fgg-hf-gg\u002Fgemma-4-26B-A4B-it","summaries\u002Fdd59ec1e4e966fad-gemma-4-26b-a4b-4b-active-moe-for-multimodal-ai-summary",[95,96,97,98],"llm","coding","agents","open-source","Gemma 4 26B A4B-it uses 26B total params but activates only 3.8B for fast inference, topping charts in reasoning (MMLU Pro 82.6%), coding (LiveCodeBench 77.1%), and vision tasks with 256K context.",[],"Cz_Nvd-bZ5BVnuCzCUIhoGq6aDorX4Dlxk8z_nNVWZQ",[103,106,109,111,114,117,119,121,123,125,127,129,132,134,136,138,140,142,144,146,148,150,153,156,158,160,163,165,167,170,172,174,176,178,180,182,184,186,188,190,192,194,196,198,200,202,204,206,208,210,212,214,216,218,220,222,224,226,228,230,232,234,236,238,240,242,244,246,248,250,252,254,256,258,260,262,264,266,268,270,272,274,276,278,280,282,284,286,288,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338,340,342,344,346,348,350,352,354,356,358,360,362,364,366,368,370,372,374,376,378,380,382,384,386,388,390,392,394,396,398,400,402,404,406,408,410,412,414,416,418,420,422,425,427,429,431,433,435,437,439,441,443,445,447,449,451,453,455,457,459,461,463,465,467,469,471,473,475,477,479,481,483,485,487,489,491,493,495,497,499,501,503,505,507,509,511,513,515,517,519,521,523,525,527,529,531,533,535,537,539,541,543,545,547,549,551,553,555,557,559,561,563,565,567,569,571,573,575,577,579,581,583,585,587,589,591,593,595,597,599,601,603,605,607,609,611,613,615,617,619,621,623,625,627,629,631,633,635,637,639,641,643,645,647,649,651,653,655,657,659,661,663,665,667,669,671,673,675,677,679,681,683,685,687,689,691,693,695,697,699,701,703,705,707,709,711,713,715,717,719,721,723,725,727,729,731,733,735,737,739,741,743,745,747,749,751,753,755,757,759,761,763,765,767,769,771,773,775,777,779,781,783,785,787,789,791,793,795,797,799,801,803,805,807,809,811,813,815,817,819,821,823,825,827,829,831,833,835,837,839,841,843,845,847,849,851,853,855,857,859,861,863,865,867,869,871,873,875,877,879,881,883,885,887,889,891,893,895,897,899,901,903,905,907,909,911,913,915,917,919,921,923,925,927,929,931,933,935,937,939,941,943,945,947,949,951,953,955,957,959,961,963,965,967,969,971,973,975,977,979,981,983,985,987,989,991,993,995,997,999,1001,1003,1005,1007,1009,1011,1013,1015,1017,1019,1021,1023,1025,1027,1029,1031,1033,1035,1037,1039,1041,1043,1045,1047,1049,1051,1053,1055,1057,1059,1061,1063,1065,1067,1069,1071,1073,1075,1077,1079,1081,1083,1085,1087,1089,1091,1093,1095,1097,1099,1101,1103,1105,1107,1109,1111,1113,1115,1117,1119,1121,1123,1125,1127,1129,1131,1133,1135,1137,1139,1141,1143,1145,1147,1149,1151,1153,1155,1157,1159,1161,1163,1165,1167,1169,1171,1173,1175,1177,1179,1181,1183,1185,1187,1189,1191,1193,1195,1197,1199,1201,1203,1205,1207,1209,1211,1213,1215,1217,1219,1221,1223,1225,1227,1229,1231,1233,1235,1237,1239,1241,1243,1245,1247,1249,1251,1253,1255,1257,1259,1261,1263,1265,1267,1269,1271,1273,1275,1277,1279,1281,1283,1285,1287,1289,1291,1293,1295,1297,1299,1301,1303,1305,1307,1309,1311,1313,1315,1317,1319,1321,1323,1325,1327,1329,1331,1333,1335,1337,1339,1341,1343,1345,1347,1349,1351,1353,1355,1357,1359,1361,1363,1365,1367,1369,1371,1373,1375,1377,1379,1381,1383,1385,1387,1389,1391,1393,1395,1397,1399,1401,1403,1405,1407,1409,1411,1413,1415,1417,1419,1421,1423,1425,1427,1429,1431,1433,1435,1437,1439,1441,1443,1445,1447,1449,1451,1453,1455,1457,1459,1461,1463,1465,1467,1469,1471,1473,1475,1477,1479,1481,1483,1485,1487,1489,1491,1493,1495,1497,1499,1501,1503,1505,1507,1509,1511,1513,1515,1517,1519,1521,1523,1525,1527,1529,1531,1533,1535,1537,1539,1541,1543,1545,1547,1549,1551,1553,1555,1557,1559,1561,1563,1565,1567,1569,1571,1573,1575,1577,1579,1581,1583,1585,1587,1589,1591,1593,1595,1597,1599,1601,1603,1605,1607,1609,1611,1613,1615,1617,1619,1621,1623,1625,1627,1629,1631,1633,1635,1637,1639,1641,1643,1645,1647,1649,1651,1653,1655,1657,1659,1661,1663,1665,1667,1669,1671,1673,1675,1677,1679,1681,1683,1685,1687,1689,1691,1693,1695,1697,1699,1701,1703,1705,1707,1709,1711,1713,1715,1717,1719,1721,1723,1725,1727,1729,1731,1733,1735,1737,1739,1741,1743,1745,1747,1749,1751,1753,1755,1757,1759,1761,1763,1765,1767,1769,1771,1773,1775,1777,1779,1781,1783,1785,1787,1789,1791,1793,1795,1797,1799,1801,1803,1805,1807,1809,1811,1813,1815,1817,1819,1821,1823,1825,1827,1829,1831,1833,1835,1837,1839,1841,1843,1845,1847,1849,1851,1853,1855,1857,1859,1861,1863,1865,1867,1869,1871,1873,1875,1877,1879,1881,1883,1885,1887,1889,1891,1893,1895,1897,1899,1901,1903,1905,1907,1909,1911,1913,1915,1917,1919,1921,1923,1925,1927,1929,1931,1933,1935,1937,1939,1941,1943,1945,1947,1949,1951,1953,1955,1957,1959,1961,1963,1965,1967,1969,1971,1973,1975,1977,1979,1981,1983,1985,1987,1989,1991,1993,1995,1997,1999,2001,2003,2005,2007,2009,2011,2013,2015,2017,2019,2021,2023,2025,2027,2029,2031,2033,2035,2037,2039,2041,2043,2045,2047,2049,2051,2053,2055,2057,2059,2061,2063,2065,2067,2069,2071,2073,2075,2077,2079,2081,2083,2085,2087,2089,2091,2093,2095,2097,2099,2101,2103,2105,2107,2109,2111,2113,2115,2117,2119,2121,2123,2125,2127,2129,2131,2133,2135,2137,2139,2141,2143,2145,2147,2149,2151,2153,2155,2157,2159,2161,2163,2165,2167,2169,2171,2173,2175,2177,2179,2181,2183,2185,2187,2189,2191,2193,2195,2197,2199,2201,2203,2205,2207,2209,2211,2213,2215,2217,2219,2221,2223,2225,2227,2229,2231,2233,2235,2237,2239,2241,2243,2245,2247,2249,2251,2253,2255,2257,2259,2261,2263,2265,2267,2269,2271,2273,2275,2277,2279,2281,2283,2285,2287,2289,2291,2293,2295,2297,2299,2301,2303,2305,2307,2309,2311,2313,2315,2317,2319,2321,2323,2325,2327,2329,2331,2333,2335,2337,2339,2341,2343,2345,2347,2349,2351,2353,2355,2357,2359,2361,2363,2365,2367,2369,2371,2373,2375,2377,2379,2381,2383,2385,2387,2389,2391,2393,2395,2397,2399,2401,2403,2405,2407,2409,2411,2413,2415,2417,2419,2421,2423,2425,2427,2429,2431,2433,2435,2437,2439,2441,2443,2445,2447,2449,2451,2453,2455,2457,2459,2461,2463,2465,2467,2469,2471,2473,2475,2477,2479,2481,2483,2485,2487,2489,2491,2493,2495,2497,2499,2501,2503,2505,2507,2509,2511,2513,2515,2517,2519,2521,2523,2525,2527,2529,2531,2533,2535,2537,2539,2541,2543,2545,2547,2549,2551,2553,2555,2557,2559,2561,2563,2565,2567,2569,2571,2573,2575,2577,2579,2581,2583,2585,2587,2589,2591,2593,2595,2597,2599,2601,2603,2605,2607,2609,2611,2613,2615,2617,2619,2621,2623,2625,2627,2629,2631,2633,2635,2637,2639,2641,2643,2645,2647,2649,2651,2653,2655,2657,2659,2661,2663,2665,2667,2669,2671,2673,2675,2677,2679,2681,2683,2685,2687,2689,2691,2693,2695,2697,2699,2701,2703,2705,2707,2709,2711,2713,2715,2717,2719,2721,2723,2725,2727,2729,2731,2733,2735,2737,2739,2741,2743,2745,2747,2749,2751,2753,2755,2757,2759,2761,2763,2765,2767,2769,2771,2773,2775,2777,2779,2781,2783,2785,2787,2789,2791,2793,2795,2797,2799,2801,2803,2805,2807,2809,2811,2813,2815,2817,2819,2821,2823,2825,2827,2829,2831,2833,2835,2837,2839,2841,2843,2845,2847,2849,2851,2853,2855,2857,2859,2861,2863,2865,2867,2869,2871,2873,2875,2877,2879,2881,2883,2885,2887,2889,2891,2893,2895,2897,2899,2901,2903,2905,2907,2909,2911,2913,2915,2917,2919,2921,2923,2925,2927,2929,2931,2933,2935,2937,2939,2941,2943,2945,2947,2949,2951,2953,2955,2957,2959,2961,2963,2965,2967,2969,2971,2973,2975,2977,2979,2981,2983,2985,2987,2989,2991,2993,2995,2997,2999,3001,3003,3005,3007,3009,3011,3013,3015,3017,3019,3021,3023,3025,3027,3029,3031,3033,3035,3037,3039,3041,3043,3045,3047,3049,3051,3053,3055,3057,3059,3061,3063,3065,3067,3069,3071,3073,3075,3077,3079,3081,3083,3085,3087,3089,3091,3093,3095,3097,3099,3101,3103,3105,3107,3109,3111,3113,3115,3117,3119,3121,3123,3125,3127,3129,3131,3133,3135,3137,3139,3141,3143,3145,3147,3149,3151,3153,3155,3157,3159,3161,3163,3165,3167,3169,3171,3173,3175,3177,3179,3181,3183,3185,3187,3189,3191,3193,3195,3197,3199,3201,3203,3205,3207,3209,3211,3213,3215,3217,3219,3221,3223,3225,3227,3229,3231,3233,3235,3237,3239,3241,3243,3245,3247,3249,3251,3253,3255,3257,3259,3261,3263,3265,3267,3269,3271,3273,3275,3277,3279,3281,3283,3285,3287,3289,3291,3293,3295,3297,3299,3301,3303,3305,3307,3309,3311,3313,3315,3317,3319,3321,3323,3325,3327,3329,3331,3333,3335,3337,3339,3341,3343,3345,3347,3349,3351,3353,3355,3357,3359,3361,3363,3365,3367,3369,3371,3373,3375,3377,3379,3381,3383,3385,3387,3389,3391,3393,3395,3397,3399,3401,3403,3405,3407,3409,3411,3413,3415,3417,3419,3421,3423,3425,3427,3429,3431,3433,3435,3437,3439,3441,3443,3445,3447,3449,3451,3453,3455,3457,3459,3461,3463,3465,3467,3469,3471,3473,3475,3477,3479,3481,3483,3485,3487,3489,3491,3493,3495,3497,3499,3501,3503,3505,3507,3509,3511,3513,3515,3517,3519,3521,3523,3525,3527,3529,3531,3533,3535,3537,3539,3541,3543,3545,3547,3549,3551,3553,3555,3557,3559,3561,3563,3565,3567,3569,3571,3573,3575,3577,3579,3581,3583,3585,3587,3589,3591,3593,3595,3597,3599,3601,3603,3605,3607,3609,3611,3613,3615,3617,3619,3621,3623,3625,3627,3629,3631,3633,3635,3637,3639,3641,3643,3645,3647,3649,3651,3653,3655,3657,3659,3661,3663,3665,3667,3669],{"categories":104},[105],"Developer Productivity",{"categories":107},[108],"Business & SaaS",{"categories":110},[55],{"categories":112},[113],"AI Automation",{"categories":115},[116],"Product Strategy",{"categories":118},[55],{"categories":120},[105],{"categories":122},[108],{"categories":124},[],{"categories":126},[55],{"categories":128},[],{"categories":130},[131],"AI News & Trends",{"categories":133},[113],{"categories":135},[131],{"categories":137},[113],{"categories":139},[113],{"categories":141},[55],{"categories":143},[55],{"categories":145},[131],{"categories":147},[55],{"categories":149},[],{"categories":151},[152],"Design & Frontend",{"categories":154},[155],"Data Science & Visualization",{"categories":157},[131],{"categories":159},[],{"categories":161},[162],"Software Engineering",{"categories":164},[55],{"categories":166},[113],{"categories":168},[169],"Marketing & Growth",{"categories":171},[55],{"categories":173},[113],{"categories":175},[],{"categories":177},[],{"categories":179},[152],{"categories":181},[113],{"categories":183},[105],{"categories":185},[152],{"categories":187},[55],{"categories":189},[113],{"categories":191},[131],{"categories":193},[],{"categories":195},[],{"categories":197},[113],{"categories":199},[162],{"categories":201},[],{"categories":203},[108],{"categories":205},[],{"categories":207},[],{"categories":209},[113],{"categories":211},[113],{"categories":213},[55],{"categories":215},[],{"categories":217},[162],{"categories":219},[],{"categories":221},[],{"categories":223},[],{"categories":225},[55],{"categories":227},[169],{"categories":229},[152],{"categories":231},[152],{"categories":233},[55],{"categories":235},[113],{"categories":237},[55],{"categories":239},[55],{"categories":241},[113],{"categories":243},[113],{"categories":245},[155],{"categories":247},[131],{"categories":249},[113],{"categories":251},[169],{"categories":253},[113],{"categories":255},[116],{"categories":257},[],{"categories":259},[113],{"categories":261},[],{"categories":263},[113],{"categories":265},[162],{"categories":267},[152],{"categories":269},[55],{"categories":271},[],{"categories":273},[],{"categories":275},[113],{"categories":277},[],{"categories":279},[55],{"categories":281},[],{"categories":283},[105],{"categories":285},[162],{"categories":287},[108],{"categories":289},[131],{"categories":291},[55],{"categories":293},[],{"categories":295},[55],{"categories":297},[],{"categories":299},[162],{"categories":301},[155],{"categories":303},[],{"categories":305},[55],{"categories":307},[152],{"categories":309},[],{"categories":311},[152],{"categories":313},[113],{"categories":315},[],{"categories":317},[113],{"categories":319},[131],{"categories":321},[55],{"categories":323},[],{"categories":325},[113],{"categories":327},[55],{"categories":329},[116],{"categories":331},[],{"categories":333},[55],{"categories":335},[113],{"categories":337},[113],{"categories":339},[],{"categories":341},[155],{"categories":343},[55],{"categories":345},[],{"categories":347},[105],{"categories":349},[108],{"categories":351},[55],{"categories":353},[113],{"categories":355},[162],{"categories":357},[55],{"categories":359},[],{"categories":361},[],{"categories":363},[55],{"categories":365},[],{"categories":367},[152],{"categories":369},[],{"categories":371},[55],{"categories":373},[],{"categories":375},[113],{"categories":377},[55],{"categories":379},[152],{"categories":381},[],{"categories":383},[55],{"categories":385},[55],{"categories":387},[108],{"categories":389},[113],{"categories":391},[55],{"categories":393},[152],{"categories":395},[113],{"categories":397},[],{"categories":399},[],{"categories":401},[131],{"categories":403},[],{"categories":405},[55],{"categories":407},[108,169],{"categories":409},[],{"categories":411},[55],{"categories":413},[],{"categories":415},[],{"categories":417},[55],{"categories":419},[],{"categories":421},[55],{"categories":423},[424],"DevOps & Cloud",{"categories":426},[],{"categories":428},[131],{"categories":430},[152],{"categories":432},[],{"categories":434},[131],{"categories":436},[131],{"categories":438},[55],{"categories":440},[169],{"categories":442},[],{"categories":444},[108],{"categories":446},[],{"categories":448},[55,424],{"categories":450},[55],{"categories":452},[55],{"categories":454},[113],{"categories":456},[55,162],{"categories":458},[155],{"categories":460},[55],{"categories":462},[169],{"categories":464},[113],{"categories":466},[113],{"categories":468},[],{"categories":470},[113],{"categories":472},[55,108],{"categories":474},[],{"categories":476},[152],{"categories":478},[152],{"categories":480},[],{"categories":482},[],{"categories":484},[131],{"categories":486},[],{"categories":488},[105],{"categories":490},[162],{"categories":492},[55],{"categories":494},[152],{"categories":496},[113],{"categories":498},[162],{"categories":500},[131],{"categories":502},[152],{"categories":504},[],{"categories":506},[55],{"categories":508},[55],{"categories":510},[55],{"categories":512},[131],{"categories":514},[105],{"categories":516},[55],{"categories":518},[113],{"categories":520},[424],{"categories":522},[152],{"categories":524},[113],{"categories":526},[],{"categories":528},[],{"categories":530},[152],{"categories":532},[131],{"categories":534},[155],{"categories":536},[],{"categories":538},[55],{"categories":540},[55],{"categories":542},[108],{"categories":544},[55],{"categories":546},[55],{"categories":548},[131],{"categories":550},[],{"categories":552},[113],{"categories":554},[162],{"categories":556},[],{"categories":558},[55],{"categories":560},[55],{"categories":562},[113],{"categories":564},[],{"categories":566},[],{"categories":568},[55],{"categories":570},[],{"categories":572},[108],{"categories":574},[113],{"categories":576},[],{"categories":578},[105],{"categories":580},[55],{"categories":582},[108],{"categories":584},[131],{"categories":586},[],{"categories":588},[],{"categories":590},[],{"categories":592},[131],{"categories":594},[131],{"categories":596},[],{"categories":598},[],{"categories":600},[108],{"categories":602},[],{"categories":604},[],{"categories":606},[105],{"categories":608},[],{"categories":610},[169],{"categories":612},[113],{"categories":614},[108],{"categories":616},[113],{"categories":618},[],{"categories":620},[116],{"categories":622},[152],{"categories":624},[162],{"categories":626},[55],{"categories":628},[113],{"categories":630},[108],{"categories":632},[55],{"categories":634},[],{"categories":636},[],{"categories":638},[162],{"categories":640},[155],{"categories":642},[116],{"categories":644},[113],{"categories":646},[55],{"categories":648},[],{"categories":650},[424],{"categories":652},[],{"categories":654},[113],{"categories":656},[],{"categories":658},[],{"categories":660},[55],{"categories":662},[152],{"categories":664},[169],{"categories":666},[113],{"categories":668},[],{"categories":670},[105],{"categories":672},[],{"categories":674},[131],{"categories":676},[55,424],{"categories":678},[131],{"categories":680},[55],{"categories":682},[108],{"categories":684},[55],{"categories":686},[],{"categories":688},[108],{"categories":690},[],{"categories":692},[162],{"categories":694},[152],{"categories":696},[131],{"categories":698},[155],{"categories":700},[105],{"categories":702},[55],{"categories":704},[162],{"categories":706},[],{"categories":708},[],{"categories":710},[116],{"categories":712},[],{"categories":714},[55],{"categories":716},[],{"categories":718},[152],{"categories":720},[152],{"categories":722},[152],{"categories":724},[],{"categories":726},[],{"categories":728},[131],{"categories":730},[113],{"categories":732},[55],{"categories":734},[55],{"categories":736},[55],{"categories":738},[108],{"categories":740},[55],{"categories":742},[],{"categories":744},[162],{"categories":746},[162],{"categories":748},[108],{"categories":750},[],{"categories":752},[55],{"categories":754},[55],{"categories":756},[108],{"categories":758},[131],{"categories":760},[169],{"categories":762},[113],{"categories":764},[],{"categories":766},[152],{"categories":768},[],{"categories":770},[55],{"categories":772},[],{"categories":774},[108],{"categories":776},[113],{"categories":778},[],{"categories":780},[424],{"categories":782},[155],{"categories":784},[162],{"categories":786},[169],{"categories":788},[162],{"categories":790},[113],{"categories":792},[],{"categories":794},[],{"categories":796},[113],{"categories":798},[105],{"categories":800},[113],{"categories":802},[116],{"categories":804},[108],{"categories":806},[],{"categories":808},[55],{"categories":810},[116],{"categories":812},[55],{"categories":814},[55],{"categories":816},[169],{"categories":818},[152],{"categories":820},[113],{"categories":822},[],{"categories":824},[],{"categories":826},[424],{"categories":828},[162],{"categories":830},[],{"categories":832},[113],{"categories":834},[55],{"categories":836},[152,55],{"categories":838},[105],{"categories":840},[],{"categories":842},[55],{"categories":844},[105],{"categories":846},[152],{"categories":848},[113],{"categories":850},[162],{"categories":852},[],{"categories":854},[55],{"categories":856},[],{"categories":858},[105],{"categories":860},[],{"categories":862},[113],{"categories":864},[116],{"categories":866},[55],{"categories":868},[55],{"categories":870},[152],{"categories":872},[113],{"categories":874},[424],{"categories":876},[152],{"categories":878},[113],{"categories":880},[55],{"categories":882},[55],{"categories":884},[55],{"categories":886},[131],{"categories":888},[],{"categories":890},[116],{"categories":892},[113],{"categories":894},[152],{"categories":896},[113],{"categories":898},[162],{"categories":900},[152],{"categories":902},[113],{"categories":904},[131],{"categories":906},[],{"categories":908},[55],{"categories":910},[152],{"categories":912},[55],{"categories":914},[105],{"categories":916},[131],{"categories":918},[55],{"categories":920},[169],{"categories":922},[55],{"categories":924},[55],{"categories":926},[113],{"categories":928},[113],{"categories":930},[55],{"categories":932},[113],{"categories":934},[152],{"categories":936},[55],{"categories":938},[],{"categories":940},[],{"categories":942},[162],{"categories":944},[],{"categories":946},[105],{"categories":948},[424],{"categories":950},[],{"categories":952},[105],{"categories":954},[108],{"categories":956},[169],{"categories":958},[],{"categories":960},[108],{"categories":962},[],{"categories":964},[],{"categories":966},[],{"categories":968},[],{"categories":970},[],{"categories":972},[55],{"categories":974},[113],{"categories":976},[424],{"categories":978},[105],{"categories":980},[55],{"categories":982},[162],{"categories":984},[116],{"categories":986},[55],{"categories":988},[169],{"categories":990},[55],{"categories":992},[55],{"categories":994},[55],{"categories":996},[55,105],{"categories":998},[162],{"categories":1000},[162],{"categories":1002},[152],{"categories":1004},[55],{"categories":1006},[],{"categories":1008},[],{"categories":1010},[],{"categories":1012},[162],{"categories":1014},[155],{"categories":1016},[131],{"categories":1018},[152],{"categories":1020},[],{"categories":1022},[55],{"categories":1024},[55],{"categories":1026},[],{"categories":1028},[],{"categories":1030},[113],{"categories":1032},[55],{"categories":1034},[108],{"categories":1036},[],{"categories":1038},[105],{"categories":1040},[55],{"categories":1042},[105],{"categories":1044},[55],{"categories":1046},[162],{"categories":1048},[169],{"categories":1050},[55,152],{"categories":1052},[131],{"categories":1054},[152],{"categories":1056},[],{"categories":1058},[424],{"categories":1060},[152],{"categories":1062},[113],{"categories":1064},[],{"categories":1066},[],{"categories":1068},[],{"categories":1070},[],{"categories":1072},[162],{"categories":1074},[113],{"categories":1076},[113],{"categories":1078},[55],{"categories":1080},[55],{"categories":1082},[],{"categories":1084},[152],{"categories":1086},[],{"categories":1088},[],{"categories":1090},[113],{"categories":1092},[],{"categories":1094},[],{"categories":1096},[169],{"categories":1098},[169],{"categories":1100},[113],{"categories":1102},[],{"categories":1104},[55],{"categories":1106},[55],{"categories":1108},[162],{"categories":1110},[152],{"categories":1112},[152],{"categories":1114},[113],{"categories":1116},[105],{"categories":1118},[55],{"categories":1120},[152],{"categories":1122},[152],{"categories":1124},[113],{"categories":1126},[113],{"categories":1128},[55],{"categories":1130},[],{"categories":1132},[],{"categories":1134},[55],{"categories":1136},[113],{"categories":1138},[131],{"categories":1140},[162],{"categories":1142},[105],{"categories":1144},[55],{"categories":1146},[],{"categories":1148},[113],{"categories":1150},[113],{"categories":1152},[],{"categories":1154},[105],{"categories":1156},[55],{"categories":1158},[105],{"categories":1160},[105],{"categories":1162},[],{"categories":1164},[],{"categories":1166},[113],{"categories":1168},[113],{"categories":1170},[55],{"categories":1172},[55],{"categories":1174},[131],{"categories":1176},[155],{"categories":1178},[116],{"categories":1180},[131],{"categories":1182},[152],{"categories":1184},[],{"categories":1186},[131],{"categories":1188},[],{"categories":1190},[],{"categories":1192},[],{"categories":1194},[],{"categories":1196},[162],{"categories":1198},[155],{"categories":1200},[],{"categories":1202},[55],{"categories":1204},[55],{"categories":1206},[155],{"categories":1208},[162],{"categories":1210},[],{"categories":1212},[],{"categories":1214},[113],{"categories":1216},[131],{"categories":1218},[131],{"categories":1220},[113],{"categories":1222},[105],{"categories":1224},[55,424],{"categories":1226},[],{"categories":1228},[152],{"categories":1230},[105],{"categories":1232},[113],{"categories":1234},[152],{"categories":1236},[],{"categories":1238},[113],{"categories":1240},[113],{"categories":1242},[55],{"categories":1244},[169],{"categories":1246},[162],{"categories":1248},[152],{"categories":1250},[],{"categories":1252},[113],{"categories":1254},[55],{"categories":1256},[113],{"categories":1258},[113],{"categories":1260},[113],{"categories":1262},[169],{"categories":1264},[113],{"categories":1266},[55],{"categories":1268},[],{"categories":1270},[169],{"categories":1272},[131],{"categories":1274},[113],{"categories":1276},[],{"categories":1278},[],{"categories":1280},[55],{"categories":1282},[113],{"categories":1284},[131],{"categories":1286},[113],{"categories":1288},[],{"categories":1290},[],{"categories":1292},[],{"categories":1294},[113],{"categories":1296},[],{"categories":1298},[],{"categories":1300},[155],{"categories":1302},[55],{"categories":1304},[155],{"categories":1306},[131],{"categories":1308},[55],{"categories":1310},[55],{"categories":1312},[113],{"categories":1314},[55],{"categories":1316},[],{"categories":1318},[],{"categories":1320},[424],{"categories":1322},[],{"categories":1324},[],{"categories":1326},[105],{"categories":1328},[],{"categories":1330},[],{"categories":1332},[],{"categories":1334},[],{"categories":1336},[162],{"categories":1338},[131],{"categories":1340},[169],{"categories":1342},[108],{"categories":1344},[55],{"categories":1346},[55],{"categories":1348},[108],{"categories":1350},[],{"categories":1352},[152],{"categories":1354},[113],{"categories":1356},[108],{"categories":1358},[55],{"categories":1360},[55],{"categories":1362},[105],{"categories":1364},[],{"categories":1366},[105],{"categories":1368},[55],{"categories":1370},[169],{"categories":1372},[113],{"categories":1374},[131],{"categories":1376},[108],{"categories":1378},[55],{"categories":1380},[113],{"categories":1382},[],{"categories":1384},[55],{"categories":1386},[105],{"categories":1388},[55],{"categories":1390},[],{"categories":1392},[131],{"categories":1394},[55],{"categories":1396},[],{"categories":1398},[108],{"categories":1400},[55],{"categories":1402},[],{"categories":1404},[],{"categories":1406},[],{"categories":1408},[55],{"categories":1410},[],{"categories":1412},[424],{"categories":1414},[55],{"categories":1416},[],{"categories":1418},[55],{"categories":1420},[55],{"categories":1422},[55],{"categories":1424},[55,424],{"categories":1426},[55],{"categories":1428},[55],{"categories":1430},[152],{"categories":1432},[113],{"categories":1434},[],{"categories":1436},[113],{"categories":1438},[55],{"categories":1440},[55],{"categories":1442},[55],{"categories":1444},[105],{"categories":1446},[105],{"categories":1448},[162],{"categories":1450},[152],{"categories":1452},[113],{"categories":1454},[],{"categories":1456},[55],{"categories":1458},[131],{"categories":1460},[55],{"categories":1462},[108],{"categories":1464},[],{"categories":1466},[424],{"categories":1468},[152],{"categories":1470},[152],{"categories":1472},[113],{"categories":1474},[131],{"categories":1476},[113],{"categories":1478},[55],{"categories":1480},[],{"categories":1482},[55],{"categories":1484},[],{"categories":1486},[],{"categories":1488},[55],{"categories":1490},[55],{"categories":1492},[55],{"categories":1494},[113],{"categories":1496},[55],{"categories":1498},[],{"categories":1500},[155],{"categories":1502},[113],{"categories":1504},[],{"categories":1506},[55],{"categories":1508},[131],{"categories":1510},[],{"categories":1512},[152],{"categories":1514},[424],{"categories":1516},[131],{"categories":1518},[162],{"categories":1520},[162],{"categories":1522},[131],{"categories":1524},[131],{"categories":1526},[424],{"categories":1528},[],{"categories":1530},[131],{"categories":1532},[55],{"categories":1534},[105],{"categories":1536},[131],{"categories":1538},[],{"categories":1540},[155],{"categories":1542},[131],{"categories":1544},[162],{"categories":1546},[131],{"categories":1548},[424],{"categories":1550},[55],{"categories":1552},[55],{"categories":1554},[],{"categories":1556},[108],{"categories":1558},[],{"categories":1560},[],{"categories":1562},[55],{"categories":1564},[55],{"categories":1566},[55],{"categories":1568},[55],{"categories":1570},[],{"categories":1572},[155],{"categories":1574},[105],{"categories":1576},[],{"categories":1578},[55],{"categories":1580},[55],{"categories":1582},[424],{"categories":1584},[424],{"categories":1586},[],{"categories":1588},[113],{"categories":1590},[131],{"categories":1592},[131],{"categories":1594},[55],{"categories":1596},[113],{"categories":1598},[],{"categories":1600},[152],{"categories":1602},[55],{"categories":1604},[55],{"categories":1606},[],{"categories":1608},[],{"categories":1610},[424],{"categories":1612},[55],{"categories":1614},[162],{"categories":1616},[108],{"categories":1618},[55],{"categories":1620},[],{"categories":1622},[113],{"categories":1624},[105],{"categories":1626},[105],{"categories":1628},[],{"categories":1630},[55],{"categories":1632},[152],{"categories":1634},[113],{"categories":1636},[],{"categories":1638},[55],{"categories":1640},[55],{"categories":1642},[113],{"categories":1644},[],{"categories":1646},[113],{"categories":1648},[162],{"categories":1650},[],{"categories":1652},[55],{"categories":1654},[],{"categories":1656},[55],{"categories":1658},[],{"categories":1660},[55],{"categories":1662},[55],{"categories":1664},[],{"categories":1666},[55],{"categories":1668},[131],{"categories":1670},[55],{"categories":1672},[55],{"categories":1674},[105],{"categories":1676},[55],{"categories":1678},[131],{"categories":1680},[113],{"categories":1682},[],{"categories":1684},[55],{"categories":1686},[169],{"categories":1688},[],{"categories":1690},[],{"categories":1692},[],{"categories":1694},[105],{"categories":1696},[131],{"categories":1698},[113],{"categories":1700},[55],{"categories":1702},[152],{"categories":1704},[113],{"categories":1706},[],{"categories":1708},[113],{"categories":1710},[],{"categories":1712},[55],{"categories":1714},[113],{"categories":1716},[55],{"categories":1718},[],{"categories":1720},[55],{"categories":1722},[55],{"categories":1724},[131],{"categories":1726},[152],{"categories":1728},[113],{"categories":1730},[152],{"categories":1732},[108],{"categories":1734},[],{"categories":1736},[],{"categories":1738},[55],{"categories":1740},[105],{"categories":1742},[131],{"categories":1744},[],{"categories":1746},[],{"categories":1748},[162],{"categories":1750},[152],{"categories":1752},[],{"categories":1754},[55],{"categories":1756},[],{"categories":1758},[169],{"categories":1760},[55],{"categories":1762},[424],{"categories":1764},[162],{"categories":1766},[],{"categories":1768},[113],{"categories":1770},[55],{"categories":1772},[113],{"categories":1774},[113],{"categories":1776},[55],{"categories":1778},[],{"categories":1780},[105],{"categories":1782},[55],{"categories":1784},[108],{"categories":1786},[162],{"categories":1788},[152],{"categories":1790},[],{"categories":1792},[],{"categories":1794},[],{"categories":1796},[113],{"categories":1798},[152],{"categories":1800},[131],{"categories":1802},[55],{"categories":1804},[131],{"categories":1806},[152],{"categories":1808},[],{"categories":1810},[152],{"categories":1812},[131],{"categories":1814},[108],{"categories":1816},[55],{"categories":1818},[131],{"categories":1820},[169],{"categories":1822},[],{"categories":1824},[],{"categories":1826},[155],{"categories":1828},[55,162],{"categories":1830},[131],{"categories":1832},[55],{"categories":1834},[113],{"categories":1836},[113],{"categories":1838},[55],{"categories":1840},[],{"categories":1842},[162],{"categories":1844},[55],{"categories":1846},[155],{"categories":1848},[113],{"categories":1850},[169],{"categories":1852},[424],{"categories":1854},[],{"categories":1856},[105],{"categories":1858},[113],{"categories":1860},[113],{"categories":1862},[162],{"categories":1864},[55],{"categories":1866},[55],{"categories":1868},[],{"categories":1870},[],{"categories":1872},[],{"categories":1874},[424],{"categories":1876},[131],{"categories":1878},[55],{"categories":1880},[55],{"categories":1882},[55],{"categories":1884},[],{"categories":1886},[155],{"categories":1888},[108],{"categories":1890},[],{"categories":1892},[113],{"categories":1894},[424],{"categories":1896},[],{"categories":1898},[152],{"categories":1900},[152],{"categories":1902},[],{"categories":1904},[162],{"categories":1906},[152],{"categories":1908},[55],{"categories":1910},[],{"categories":1912},[131],{"categories":1914},[55],{"categories":1916},[152],{"categories":1918},[113],{"categories":1920},[131],{"categories":1922},[],{"categories":1924},[113],{"categories":1926},[152],{"categories":1928},[55],{"categories":1930},[],{"categories":1932},[55],{"categories":1934},[55],{"categories":1936},[424],{"categories":1938},[131],{"categories":1940},[155],{"categories":1942},[155],{"categories":1944},[],{"categories":1946},[],{"categories":1948},[],{"categories":1950},[113],{"categories":1952},[162],{"categories":1954},[162],{"categories":1956},[],{"categories":1958},[],{"categories":1960},[55],{"categories":1962},[],{"categories":1964},[113],{"categories":1966},[55],{"categories":1968},[],{"categories":1970},[55],{"categories":1972},[108],{"categories":1974},[55],{"categories":1976},[169],{"categories":1978},[113],{"categories":1980},[55],{"categories":1982},[162],{"categories":1984},[131],{"categories":1986},[113],{"categories":1988},[],{"categories":1990},[131],{"categories":1992},[113],{"categories":1994},[113],{"categories":1996},[],{"categories":1998},[108],{"categories":2000},[113],{"categories":2002},[],{"categories":2004},[55],{"categories":2006},[105],{"categories":2008},[131],{"categories":2010},[424],{"categories":2012},[113],{"categories":2014},[113],{"categories":2016},[105],{"categories":2018},[55],{"categories":2020},[],{"categories":2022},[],{"categories":2024},[152],{"categories":2026},[55,108],{"categories":2028},[],{"categories":2030},[105],{"categories":2032},[155],{"categories":2034},[55],{"categories":2036},[162],{"categories":2038},[55],{"categories":2040},[113],{"categories":2042},[55],{"categories":2044},[55],{"categories":2046},[131],{"categories":2048},[113],{"categories":2050},[],{"categories":2052},[],{"categories":2054},[113],{"categories":2056},[55],{"categories":2058},[424],{"categories":2060},[],{"categories":2062},[55],{"categories":2064},[113],{"categories":2066},[],{"categories":2068},[55],{"categories":2070},[169],{"categories":2072},[155],{"categories":2074},[113],{"categories":2076},[55],{"categories":2078},[424],{"categories":2080},[],{"categories":2082},[55],{"categories":2084},[169],{"categories":2086},[152],{"categories":2088},[55],{"categories":2090},[],{"categories":2092},[169],{"categories":2094},[131],{"categories":2096},[55],{"categories":2098},[55],{"categories":2100},[105],{"categories":2102},[],{"categories":2104},[],{"categories":2106},[152],{"categories":2108},[55],{"categories":2110},[155],{"categories":2112},[169],{"categories":2114},[169],{"categories":2116},[131],{"categories":2118},[],{"categories":2120},[],{"categories":2122},[55],{"categories":2124},[],{"categories":2126},[55,162],{"categories":2128},[131],{"categories":2130},[113],{"categories":2132},[162],{"categories":2134},[55],{"categories":2136},[105],{"categories":2138},[],{"categories":2140},[],{"categories":2142},[105],{"categories":2144},[169],{"categories":2146},[55],{"categories":2148},[],{"categories":2150},[152,55],{"categories":2152},[424],{"categories":2154},[105],{"categories":2156},[],{"categories":2158},[108],{"categories":2160},[108],{"categories":2162},[55],{"categories":2164},[162],{"categories":2166},[113],{"categories":2168},[131],{"categories":2170},[169],{"categories":2172},[152],{"categories":2174},[55],{"categories":2176},[55],{"categories":2178},[55],{"categories":2180},[105],{"categories":2182},[55],{"categories":2184},[113],{"categories":2186},[131],{"categories":2188},[],{"categories":2190},[],{"categories":2192},[155],{"categories":2194},[162],{"categories":2196},[55],{"categories":2198},[152],{"categories":2200},[155],{"categories":2202},[55],{"categories":2204},[55],{"categories":2206},[113],{"categories":2208},[113],{"categories":2210},[55,108],{"categories":2212},[],{"categories":2214},[152],{"categories":2216},[],{"categories":2218},[55],{"categories":2220},[131],{"categories":2222},[105],{"categories":2224},[105],{"categories":2226},[113],{"categories":2228},[55],{"categories":2230},[108],{"categories":2232},[162],{"categories":2234},[169],{"categories":2236},[],{"categories":2238},[131],{"categories":2240},[55],{"categories":2242},[55],{"categories":2244},[131],{"categories":2246},[162],{"categories":2248},[55],{"categories":2250},[113],{"categories":2252},[131],{"categories":2254},[55],{"categories":2256},[152],{"categories":2258},[55],{"categories":2260},[55],{"categories":2262},[424],{"categories":2264},[116],{"categories":2266},[113],{"categories":2268},[55],{"categories":2270},[131],{"categories":2272},[113],{"categories":2274},[169],{"categories":2276},[55],{"categories":2278},[],{"categories":2280},[55],{"categories":2282},[],{"categories":2284},[],{"categories":2286},[],{"categories":2288},[108],{"categories":2290},[55],{"categories":2292},[113],{"categories":2294},[131],{"categories":2296},[131],{"categories":2298},[131],{"categories":2300},[131],{"categories":2302},[],{"categories":2304},[105],{"categories":2306},[113],{"categories":2308},[131],{"categories":2310},[105],{"categories":2312},[113],{"categories":2314},[55],{"categories":2316},[55,113],{"categories":2318},[113],{"categories":2320},[424],{"categories":2322},[131],{"categories":2324},[131],{"categories":2326},[113],{"categories":2328},[55],{"categories":2330},[],{"categories":2332},[131],{"categories":2334},[169],{"categories":2336},[105],{"categories":2338},[55],{"categories":2340},[55],{"categories":2342},[],{"categories":2344},[162],{"categories":2346},[],{"categories":2348},[105],{"categories":2350},[113],{"categories":2352},[131],{"categories":2354},[55],{"categories":2356},[131],{"categories":2358},[105],{"categories":2360},[131],{"categories":2362},[131],{"categories":2364},[],{"categories":2366},[108],{"categories":2368},[113],{"categories":2370},[131],{"categories":2372},[131],{"categories":2374},[131],{"categories":2376},[131],{"categories":2378},[131],{"categories":2380},[131],{"categories":2382},[131],{"categories":2384},[131],{"categories":2386},[131],{"categories":2388},[131],{"categories":2390},[155],{"categories":2392},[105],{"categories":2394},[55],{"categories":2396},[55],{"categories":2398},[],{"categories":2400},[55,105],{"categories":2402},[],{"categories":2404},[113],{"categories":2406},[131],{"categories":2408},[113],{"categories":2410},[55],{"categories":2412},[55],{"categories":2414},[55],{"categories":2416},[55],{"categories":2418},[55],{"categories":2420},[113],{"categories":2422},[108],{"categories":2424},[152],{"categories":2426},[131],{"categories":2428},[55],{"categories":2430},[],{"categories":2432},[],{"categories":2434},[113],{"categories":2436},[152],{"categories":2438},[55],{"categories":2440},[],{"categories":2442},[],{"categories":2444},[169],{"categories":2446},[55],{"categories":2448},[],{"categories":2450},[],{"categories":2452},[105],{"categories":2454},[108],{"categories":2456},[55],{"categories":2458},[108],{"categories":2460},[152],{"categories":2462},[],{"categories":2464},[131],{"categories":2466},[],{"categories":2468},[152],{"categories":2470},[55],{"categories":2472},[169],{"categories":2474},[],{"categories":2476},[169],{"categories":2478},[],{"categories":2480},[],{"categories":2482},[113],{"categories":2484},[],{"categories":2486},[108],{"categories":2488},[105],{"categories":2490},[152],{"categories":2492},[162],{"categories":2494},[],{"categories":2496},[],{"categories":2498},[55],{"categories":2500},[105],{"categories":2502},[169],{"categories":2504},[],{"categories":2506},[113],{"categories":2508},[113],{"categories":2510},[131],{"categories":2512},[55],{"categories":2514},[113],{"categories":2516},[55],{"categories":2518},[113],{"categories":2520},[55],{"categories":2522},[116],{"categories":2524},[131],{"categories":2526},[],{"categories":2528},[169],{"categories":2530},[162],{"categories":2532},[113],{"categories":2534},[],{"categories":2536},[55],{"categories":2538},[113],{"categories":2540},[108],{"categories":2542},[105],{"categories":2544},[55],{"categories":2546},[152],{"categories":2548},[162],{"categories":2550},[162],{"categories":2552},[55],{"categories":2554},[155],{"categories":2556},[55],{"categories":2558},[113],{"categories":2560},[108],{"categories":2562},[113],{"categories":2564},[55],{"categories":2566},[55],{"categories":2568},[113],{"categories":2570},[131],{"categories":2572},[],{"categories":2574},[105],{"categories":2576},[55],{"categories":2578},[113],{"categories":2580},[55],{"categories":2582},[55],{"categories":2584},[],{"categories":2586},[152],{"categories":2588},[108],{"categories":2590},[131],{"categories":2592},[55],{"categories":2594},[55],{"categories":2596},[152],{"categories":2598},[169],{"categories":2600},[155],{"categories":2602},[55],{"categories":2604},[131],{"categories":2606},[55],{"categories":2608},[113],{"categories":2610},[424],{"categories":2612},[55],{"categories":2614},[113],{"categories":2616},[155],{"categories":2618},[],{"categories":2620},[113],{"categories":2622},[162],{"categories":2624},[152],{"categories":2626},[55],{"categories":2628},[105],{"categories":2630},[108],{"categories":2632},[162],{"categories":2634},[],{"categories":2636},[113],{"categories":2638},[55],{"categories":2640},[],{"categories":2642},[131],{"categories":2644},[],{"categories":2646},[131],{"categories":2648},[55],{"categories":2650},[113],{"categories":2652},[113],{"categories":2654},[113],{"categories":2656},[],{"categories":2658},[],{"categories":2660},[55],{"categories":2662},[55],{"categories":2664},[],{"categories":2666},[152],{"categories":2668},[113],{"categories":2670},[169],{"categories":2672},[105],{"categories":2674},[],{"categories":2676},[],{"categories":2678},[131],{"categories":2680},[162],{"categories":2682},[55],{"categories":2684},[55],{"categories":2686},[55],{"categories":2688},[162],{"categories":2690},[131],{"categories":2692},[152],{"categories":2694},[55],{"categories":2696},[55],{"categories":2698},[55],{"categories":2700},[131],{"categories":2702},[55],{"categories":2704},[131],{"categories":2706},[113],{"categories":2708},[113],{"categories":2710},[162],{"categories":2712},[113],{"categories":2714},[55],{"categories":2716},[162],{"categories":2718},[152],{"categories":2720},[],{"categories":2722},[113],{"categories":2724},[],{"categories":2726},[],{"categories":2728},[108],{"categories":2730},[55],{"categories":2732},[113],{"categories":2734},[105],{"categories":2736},[113],{"categories":2738},[169],{"categories":2740},[],{"categories":2742},[113],{"categories":2744},[],{"categories":2746},[105],{"categories":2748},[113],{"categories":2750},[],{"categories":2752},[113],{"categories":2754},[55],{"categories":2756},[131],{"categories":2758},[55],{"categories":2760},[113],{"categories":2762},[131],{"categories":2764},[113],{"categories":2766},[162],{"categories":2768},[152],{"categories":2770},[105],{"categories":2772},[],{"categories":2774},[113],{"categories":2776},[152],{"categories":2778},[131],{"categories":2780},[55],{"categories":2782},[152],{"categories":2784},[105],{"categories":2786},[],{"categories":2788},[113],{"categories":2790},[113],{"categories":2792},[55],{"categories":2794},[],{"categories":2796},[113],{"categories":2798},[116],{"categories":2800},[131],{"categories":2802},[113],{"categories":2804},[108],{"categories":2806},[],{"categories":2808},[55],{"categories":2810},[116],{"categories":2812},[55],{"categories":2814},[113],{"categories":2816},[131],{"categories":2818},[105],{"categories":2820},[424],{"categories":2822},[55],{"categories":2824},[55],{"categories":2826},[55],{"categories":2828},[131],{"categories":2830},[108],{"categories":2832},[55],{"categories":2834},[152],{"categories":2836},[131],{"categories":2838},[424],{"categories":2840},[55],{"categories":2842},[],{"categories":2844},[],{"categories":2846},[424],{"categories":2848},[155],{"categories":2850},[113],{"categories":2852},[113],{"categories":2854},[131],{"categories":2856},[55],{"categories":2858},[105],{"categories":2860},[152],{"categories":2862},[113],{"categories":2864},[55],{"categories":2866},[169],{"categories":2868},[55],{"categories":2870},[113],{"categories":2872},[],{"categories":2874},[55],{"categories":2876},[55],{"categories":2878},[131],{"categories":2880},[105],{"categories":2882},[],{"categories":2884},[55],{"categories":2886},[55],{"categories":2888},[162],{"categories":2890},[152],{"categories":2892},[55,113],{"categories":2894},[169,108],{"categories":2896},[55],{"categories":2898},[],{"categories":2900},[113],{"categories":2902},[],{"categories":2904},[162],{"categories":2906},[55],{"categories":2908},[131],{"categories":2910},[],{"categories":2912},[113],{"categories":2914},[],{"categories":2916},[113],{"categories":2918},[105],{"categories":2920},[113],{"categories":2922},[55],{"categories":2924},[424],{"categories":2926},[169],{"categories":2928},[108],{"categories":2930},[108],{"categories":2932},[105],{"categories":2934},[105],{"categories":2936},[55],{"categories":2938},[113],{"categories":2940},[55],{"categories":2942},[55],{"categories":2944},[105],{"categories":2946},[55],{"categories":2948},[169],{"categories":2950},[131],{"categories":2952},[55],{"categories":2954},[113],{"categories":2956},[55],{"categories":2958},[],{"categories":2960},[162],{"categories":2962},[],{"categories":2964},[113],{"categories":2966},[105],{"categories":2968},[],{"categories":2970},[424],{"categories":2972},[55],{"categories":2974},[],{"categories":2976},[131],{"categories":2978},[113],{"categories":2980},[162],{"categories":2982},[55],{"categories":2984},[113],{"categories":2986},[162],{"categories":2988},[113],{"categories":2990},[131],{"categories":2992},[105],{"categories":2994},[131],{"categories":2996},[162],{"categories":2998},[55],{"categories":3000},[152],{"categories":3002},[55],{"categories":3004},[55],{"categories":3006},[55],{"categories":3008},[55],{"categories":3010},[113],{"categories":3012},[55],{"categories":3014},[113],{"categories":3016},[55],{"categories":3018},[105],{"categories":3020},[55],{"categories":3022},[113],{"categories":3024},[152],{"categories":3026},[105],{"categories":3028},[113],{"categories":3030},[152],{"categories":3032},[],{"categories":3034},[55],{"categories":3036},[55],{"categories":3038},[162],{"categories":3040},[],{"categories":3042},[113],{"categories":3044},[169],{"categories":3046},[55],{"categories":3048},[131],{"categories":3050},[169],{"categories":3052},[113],{"categories":3054},[108],{"categories":3056},[108],{"categories":3058},[55],{"categories":3060},[105],{"categories":3062},[],{"categories":3064},[55],{"categories":3066},[],{"categories":3068},[105],{"categories":3070},[55],{"categories":3072},[113],{"categories":3074},[113],{"categories":3076},[],{"categories":3078},[162],{"categories":3080},[162],{"categories":3082},[169],{"categories":3084},[152],{"categories":3086},[],{"categories":3088},[55],{"categories":3090},[105],{"categories":3092},[55],{"categories":3094},[162],{"categories":3096},[105],{"categories":3098},[131],{"categories":3100},[131],{"categories":3102},[],{"categories":3104},[131],{"categories":3106},[113],{"categories":3108},[152],{"categories":3110},[155],{"categories":3112},[55],{"categories":3114},[],{"categories":3116},[131],{"categories":3118},[162],{"categories":3120},[108],{"categories":3122},[55],{"categories":3124},[105],{"categories":3126},[424],{"categories":3128},[105],{"categories":3130},[],{"categories":3132},[],{"categories":3134},[131],{"categories":3136},[],{"categories":3138},[113],{"categories":3140},[113],{"categories":3142},[113],{"categories":3144},[],{"categories":3146},[55],{"categories":3148},[],{"categories":3150},[131],{"categories":3152},[105],{"categories":3154},[152],{"categories":3156},[55],{"categories":3158},[131],{"categories":3160},[131],{"categories":3162},[],{"categories":3164},[131],{"categories":3166},[105],{"categories":3168},[55],{"categories":3170},[],{"categories":3172},[113],{"categories":3174},[113],{"categories":3176},[105],{"categories":3178},[],{"categories":3180},[],{"categories":3182},[],{"categories":3184},[152],{"categories":3186},[113],{"categories":3188},[55],{"categories":3190},[],{"categories":3192},[],{"categories":3194},[],{"categories":3196},[152],{"categories":3198},[],{"categories":3200},[105],{"categories":3202},[],{"categories":3204},[],{"categories":3206},[152],{"categories":3208},[55],{"categories":3210},[131],{"categories":3212},[],{"categories":3214},[169],{"categories":3216},[131],{"categories":3218},[169],{"categories":3220},[55],{"categories":3222},[],{"categories":3224},[],{"categories":3226},[113],{"categories":3228},[],{"categories":3230},[],{"categories":3232},[113],{"categories":3234},[55],{"categories":3236},[],{"categories":3238},[113],{"categories":3240},[131],{"categories":3242},[169],{"categories":3244},[155],{"categories":3246},[113],{"categories":3248},[113],{"categories":3250},[],{"categories":3252},[],{"categories":3254},[],{"categories":3256},[131],{"categories":3258},[],{"categories":3260},[],{"categories":3262},[152],{"categories":3264},[105],{"categories":3266},[],{"categories":3268},[108],{"categories":3270},[169],{"categories":3272},[55],{"categories":3274},[162],{"categories":3276},[105],{"categories":3278},[155],{"categories":3280},[108],{"categories":3282},[162],{"categories":3284},[],{"categories":3286},[],{"categories":3288},[113],{"categories":3290},[105],{"categories":3292},[152],{"categories":3294},[105],{"categories":3296},[113],{"categories":3298},[424],{"categories":3300},[113],{"categories":3302},[],{"categories":3304},[55],{"categories":3306},[131],{"categories":3308},[162],{"categories":3310},[],{"categories":3312},[152],{"categories":3314},[131],{"categories":3316},[105],{"categories":3318},[113],{"categories":3320},[55],{"categories":3322},[108],{"categories":3324},[113,424],{"categories":3326},[113],{"categories":3328},[162],{"categories":3330},[55],{"categories":3332},[155],{"categories":3334},[169],{"categories":3336},[113],{"categories":3338},[],{"categories":3340},[113],{"categories":3342},[55],{"categories":3344},[108],{"categories":3346},[],{"categories":3348},[],{"categories":3350},[55],{"categories":3352},[155],{"categories":3354},[55],{"categories":3356},[],{"categories":3358},[131],{"categories":3360},[],{"categories":3362},[131],{"categories":3364},[162],{"categories":3366},[113],{"categories":3368},[55],{"categories":3370},[169],{"categories":3372},[162],{"categories":3374},[],{"categories":3376},[131],{"categories":3378},[55],{"categories":3380},[],{"categories":3382},[55],{"categories":3384},[113],{"categories":3386},[55],{"categories":3388},[113],{"categories":3390},[55],{"categories":3392},[55],{"categories":3394},[55],{"categories":3396},[55],{"categories":3398},[108],{"categories":3400},[],{"categories":3402},[116],{"categories":3404},[131],{"categories":3406},[55],{"categories":3408},[],{"categories":3410},[162],{"categories":3412},[55],{"categories":3414},[55],{"categories":3416},[113],{"categories":3418},[131],{"categories":3420},[55],{"categories":3422},[55],{"categories":3424},[108],{"categories":3426},[113],{"categories":3428},[152],{"categories":3430},[],{"categories":3432},[155],{"categories":3434},[55],{"categories":3436},[],{"categories":3438},[131],{"categories":3440},[169],{"categories":3442},[],{"categories":3444},[],{"categories":3446},[131],{"categories":3448},[131],{"categories":3450},[169],{"categories":3452},[105],{"categories":3454},[113],{"categories":3456},[113],{"categories":3458},[55],{"categories":3460},[108],{"categories":3462},[],{"categories":3464},[],{"categories":3466},[131],{"categories":3468},[155],{"categories":3470},[162],{"categories":3472},[113],{"categories":3474},[152],{"categories":3476},[155],{"categories":3478},[155],{"categories":3480},[],{"categories":3482},[131],{"categories":3484},[55],{"categories":3486},[55],{"categories":3488},[162],{"categories":3490},[],{"categories":3492},[131],{"categories":3494},[131],{"categories":3496},[131],{"categories":3498},[],{"categories":3500},[113],{"categories":3502},[55],{"categories":3504},[],{"categories":3506},[105],{"categories":3508},[108],{"categories":3510},[],{"categories":3512},[55],{"categories":3514},[55],{"categories":3516},[],{"categories":3518},[162],{"categories":3520},[],{"categories":3522},[],{"categories":3524},[],{"categories":3526},[],{"categories":3528},[55],{"categories":3530},[131],{"categories":3532},[],{"categories":3534},[],{"categories":3536},[55],{"categories":3538},[55],{"categories":3540},[55],{"categories":3542},[155],{"categories":3544},[55],{"categories":3546},[155],{"categories":3548},[],{"categories":3550},[155],{"categories":3552},[155],{"categories":3554},[424],{"categories":3556},[113],{"categories":3558},[162],{"categories":3560},[],{"categories":3562},[],{"categories":3564},[155],{"categories":3566},[162],{"categories":3568},[162],{"categories":3570},[162],{"categories":3572},[],{"categories":3574},[105],{"categories":3576},[162],{"categories":3578},[162],{"categories":3580},[105],{"categories":3582},[162],{"categories":3584},[108],{"categories":3586},[162],{"categories":3588},[162],{"categories":3590},[162],{"categories":3592},[155],{"categories":3594},[131],{"categories":3596},[131],{"categories":3598},[55],{"categories":3600},[162],{"categories":3602},[155],{"categories":3604},[424],{"categories":3606},[155],{"categories":3608},[155],{"categories":3610},[155],{"categories":3612},[],{"categories":3614},[108],{"categories":3616},[],{"categories":3618},[424],{"categories":3620},[162],{"categories":3622},[162],{"categories":3624},[162],{"categories":3626},[113],{"categories":3628},[131,108],{"categories":3630},[155],{"categories":3632},[],{"categories":3634},[],{"categories":3636},[155],{"categories":3638},[],{"categories":3640},[155],{"categories":3642},[131],{"categories":3644},[113],{"categories":3646},[],{"categories":3648},[162],{"categories":3650},[55],{"categories":3652},[152],{"categories":3654},[],{"categories":3656},[55],{"categories":3658},[],{"categories":3660},[131],{"categories":3662},[105],{"categories":3664},[155],{"categories":3666},[],{"categories":3668},[162],{"categories":3670},[131],[3672,3753,3814,3953],{"id":3673,"title":3674,"ai":3675,"body":3680,"categories":3720,"created_at":56,"date_modified":56,"description":47,"extension":57,"faq":56,"featured":58,"kicker_label":56,"meta":3721,"navigation":84,"path":3742,"published_at":56,"question":56,"scraped_at":3743,"seo":3744,"sitemap":3745,"source_id":3746,"source_name":90,"source_type":91,"source_url":3747,"stem":3748,"tags":3749,"thumbnail_url":56,"tldr":3750,"tweet":56,"unknown_tags":3751,"__hash__":3752},"summaries\u002Fsummaries\u002F661d1f4b66cfe71b-glm-5-leads-open-source-in-coding-reasoning-agents-summary.md","GLM-5 Leads Open-Source in Coding, Reasoning, Agents",{"provider":7,"model":8,"input_tokens":3676,"output_tokens":3677,"processing_time_ms":3678,"cost_usd":3679},7657,1725,8371,0.00188705,{"type":14,"value":3681,"toc":3715},[3682,3686,3689,3692,3695,3699,3702,3705,3709,3712],[17,3683,3685],{"id":3684},"scale-pre-training-and-rl-for-superior-competence","Scale Pre-Training and RL for Superior Competence",[22,3687,3688],{},"GLM-5 advances from GLM-4.5's 355B params (32B active) and 23T tokens to 744B params (40B active) with 28.5T pre-training tokens, using DeepSeek Sparse Attention (DSA) to maintain long-context capacity at lower deployment costs. Post-training leverages slime, an asynchronous RL infrastructure (github.com\u002FTHUDM\u002Fslime), to boost throughput for fine-grained iterations, bridging pre-trained competence to excellence without RL's typical inefficiency.",[22,3690,3691],{},"This yields best-in-class open-source results: Humanity's Last Exam (30.5, 50.4 w\u002Ftools), AIME 2026 I (92.7%), HMMT Nov 2025 (96.9%), GPQA-Diamond (86.0%). Coding excels on SWE-bench Verified (77.8%), Multilingual (73.3%), Terminal-Bench 2.0 (56.2\u002F60.7 verified), CyberGym (43.2%). Agent benchmarks include BrowseComp (62.0\u002F75.9 w\u002Fcontext mgmt), τ²-Bench (89.7%), MCP-Atlas (67.8%), Tool-Decathlon (39.2), nearing Claude Opus 4.5 and Gemini 3.0 Pro.",[22,3693,3694],{},"On CC-Bench-V2 internal suite, GLM-5 outperforms GLM-4.7 in frontend, backend, long-horizon tasks, closing gap to Claude Opus 4.5. Vending Bench 2 simulates year-long vending machine ops; GLM-5 ends with $4,432 balance (#1 open-source, near Claude's $4,967).",[17,3696,3698],{"id":3697},"agentic-engineering-for-real-deliverables","Agentic Engineering for Real Deliverables",[22,3700,3701],{},"GLM-5 shifts from 'vibe coding' to agentic systems engineering, handling complex, multi-turn tasks like generating production-ready .docx\u002F.pdf\u002F.xlsx files (PRDs, spreadsheets, reports) from prompts. Example: Student council football sponsorship proposal includes structured sections (intro, event details, fund use, tiers, benefits), visual elements (image placeholders, tables), color palette (navy\u002Fgold), ensuring print-ready output without edits.",[22,3703,3704],{},"Z.ai's Agent mode supports PDF\u002FWord\u002FExcel creation, multi-turn collaboration for deliverables. Outperforms priors on long-horizon planning\u002Fresource mgmt, as in Vending Bench 2's operational simulation.",[17,3706,3708],{"id":3707},"deploy-anywhere-integrate-seamlessly","Deploy Anywhere, Integrate Seamlessly",[22,3710,3711],{},"Open-source (MIT) on HuggingFace (zai-org\u002FGLM-5), ModelScope, GitHub (zai-org\u002FGLM-5). API via api.z.ai, BigModel.cn, Z.ai chat\u002Fagent modes. Local inference: vLLM\u002FSGLang; non-NVIDIA chips (Ascend, Moore Threads) via quantization.",[22,3713,3714],{},"Coding agents: Claude Code\u002FOpenClaw (update to 'GLM-5', higher quota use), OpenCode\u002FKilo\u002FRoo\u002FCline\u002FDroid. Z Code IDE for multi-agent collab. Gradual rollout to GLM Coding Plan subs; free trial on Z.ai.",{"title":47,"searchDepth":48,"depth":48,"links":3716},[3717,3718,3719],{"id":3684,"depth":48,"text":3685},{"id":3697,"depth":48,"text":3698},{"id":3707,"depth":48,"text":3708},[],{"content_references":3722,"triage":3738},[3723,3727,3731,3735],{"type":3724,"title":3725,"url":3726,"context":65},"paper","GLM-5 Tech Report","https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.15763",{"type":3728,"title":3729,"url":3730,"context":65},"tool","slime","https:\u002F\u002Fgithub.com\u002FTHUDM\u002Fslime",{"type":3732,"title":3733,"url":3734,"context":65},"dataset","Terminal-Bench 2.0 Verified","https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fzai-org\u002Fterminal-bench-2-verified",{"type":62,"title":3736,"url":3737,"context":65},"Vending Bench 2","https:\u002F\u002Fandonlabs.com\u002Fevals\u002Fvending-bench-2",{"relevance":81,"novelty":3739,"quality":81,"actionability":3739,"composite":3740,"reasoning":3741},3,3.6,"Category: AI & LLMs. The article discusses GLM-5's advancements in coding and reasoning capabilities, which directly relates to AI engineering and the development of AI-powered products. It provides insights into the model's performance benchmarks and practical applications, such as generating production-ready documents, which can help developers understand how to leverage these capabilities in their projects.","\u002Fsummaries\u002F661d1f4b66cfe71b-glm-5-leads-open-source-in-coding-reasoning-agents-summary","2026-04-16 03:07:03",{"title":3674,"description":47},{"loc":3742},"661d1f4b66cfe71b","https:\u002F\u002Fz.ai\u002Fblog\u002Fglm-5","summaries\u002F661d1f4b66cfe71b-glm-5-leads-open-source-in-coding-reasoning-agents-summary",[95,97,96,98],"GLM-5 scales to 744B params (40B active) and 28.5T tokens, tops open-source benchmarks like SWE-bench (77.8%) and Vending Bench 2 ($4,432 balance), enabling complex engineering and long-horizon agents while cutting deployment costs via DSA.",[],"EVOJLevCvMfkN4EQZAjuWO1tjnTop9-wKWMYLgh9bSg",{"id":3754,"title":3755,"ai":3756,"body":3761,"categories":3789,"created_at":56,"date_modified":56,"description":47,"extension":57,"faq":56,"featured":58,"kicker_label":56,"meta":3790,"navigation":84,"path":3801,"published_at":3802,"question":56,"scraped_at":3803,"seo":3804,"sitemap":3805,"source_id":3806,"source_name":3807,"source_type":91,"source_url":3808,"stem":3809,"tags":3810,"thumbnail_url":56,"tldr":3811,"tweet":56,"unknown_tags":3812,"__hash__":3813},"summaries\u002Fsummaries\u002F138f159d6a0dc547-tokenspeed-beats-tensorrt-llm-9-11-on-agentic-codi-summary.md","TokenSpeed Beats TensorRT-LLM 9-11% on Agentic Coding Inference",{"provider":7,"model":8,"input_tokens":3757,"output_tokens":3758,"processing_time_ms":3759,"cost_usd":3760},6300,1652,21300,0.00157915,{"type":14,"value":3762,"toc":3784},[3763,3767,3770,3774,3777,3781],[17,3764,3766],{"id":3765},"tackling-agentic-inference-bottlenecks","Tackling Agentic Inference Bottlenecks",[22,3768,3769],{},"Agentic coding systems like Claude Code, Codex, and Cursor push inference engines with contexts over 50K tokens across dozens of turns, stressing per-GPU tokens-per-minute (TPM) for multi-user scaling and per-user tokens-per-second (TPS) for responsiveness (target floor: 70 TPS, up to 200+ TPS). Public benchmarks miss this dual pressure, so TokenSpeed (MIT-licensed preview from LightSeek Foundation) prioritizes both metrics via specialized architecture, avoiding generic chat optimizations.",[17,3771,3773],{"id":3772},"architectural-edges-for-speed-and-safety","Architectural Edges for Speed and Safety",[22,3775,3776],{},"TokenSpeed builds on five subsystems: (1) Compiler-backed SPMD modeling auto-generates collective ops from I\u002FO annotations, skipping manual comms code. (2) Scheduler splits C++ control plane (FSM with type-enforced KV cache ownership\u002Ftransfers for compile-time safety) from Python execution plane (fast iteration). (3) Pluggable kernel layer with registry supports heterogeneous accelerators; its MLA kernel (grouping q_seqlen\u002Fnum_heads for Tensor Core fill, tuned binary prefill softmax) beats TensorRT-LLM decode\u002Fprefill, adopted by vLLM. (4) Safe KV reuse restrictions. (5) SMG for low-overhead CPU-GPU handoff. These cut KV errors (common pitfall) and enable modular accel support beyond NVIDIA.",[17,3778,3780],{"id":3779},"benchmark-dominance-on-real-workloads","Benchmark Dominance on Real Workloads",[22,3782,3783],{},"On NVIDIA B200 with SWE-smith traces (production-like coding agent traffic) and Kimi K2.5 model, TokenSpeed in Attention TP4 + MoE TP4 config tops TensorRT-LLM Pareto: 9% faster at batch=1 min-latency (>70 TPS\u002Fuser), 11% higher throughput at ~100 TPS\u002Fuser. Decode MLA folds query-seq into head axis for better BMM tile fill; binary prefill tunes softmax. With speculative decoding + long prefix KV at batches 4\u002F8\u002F16, latency nearly halves vs. TensorRT-LLM. Single-node only for now; PD disagg coming.",{"title":47,"searchDepth":48,"depth":48,"links":3785},[3786,3787,3788],{"id":3765,"depth":48,"text":3766},{"id":3772,"depth":48,"text":3773},{"id":3779,"depth":48,"text":3780},[55],{"content_references":3791,"triage":3799},[3792,3795],{"type":3728,"title":3793,"url":3794,"context":65},"TokenSpeed","https:\u002F\u002Fgithub.com\u002Flightseekorg\u002Ftokenspeed",{"type":62,"title":3796,"url":3797,"context":3798},"LightSeek TokenSpeed Technical Details","https:\u002F\u002Flightseek.org\u002Fblog\u002Flightseek-tokenspeed.html","recommended",{"relevance":81,"novelty":3739,"quality":81,"actionability":3739,"composite":3740,"reasoning":3800},"Category: AI & LLMs. The article discusses a new open-source LLM inference engine, TokenSpeed, which addresses specific performance issues in agentic workloads, directly relevant to AI engineers and developers. It provides insights into architectural improvements and benchmarks, but lacks detailed implementation guidance for practical application.","\u002Fsummaries\u002F138f159d6a0dc547-tokenspeed-beats-tensorrt-llm-9-11-on-agentic-codi-summary","2026-05-07 22:03:47","2026-05-08 11:28:23",{"title":3755,"description":47},{"loc":3801},"138f159d6a0dc547","MarkTechPost","https:\u002F\u002Fwww.marktechpost.com\u002F2026\u002F05\u002F07\u002Flightseek-foundation-releases-tokenspeed-an-open-source-llm-inference-engine-targeting-tensorrt-llm-level-performance-for-agentic-workloads\u002F","summaries\u002F138f159d6a0dc547-tokenspeed-beats-tensorrt-llm-9-11-on-agentic-codi-summary",[95,97,98],"TokenSpeed open-source engine optimizes agentic workloads with long contexts (>50K tokens) and multi-turn convos, delivering 9% lower latency and 11% higher throughput than TensorRT-LLM at 70-100 TPS\u002Fuser on NVIDIA B200.",[],"iEELj0BdlJHttBF4chi4fvtEa11ZZu5n0Lg0DMzVIIg",{"id":3815,"title":3816,"ai":3817,"body":3822,"categories":3926,"created_at":56,"date_modified":56,"description":47,"extension":57,"faq":56,"featured":58,"kicker_label":56,"meta":3927,"navigation":84,"path":3941,"published_at":3942,"question":56,"scraped_at":3943,"seo":3944,"sitemap":3945,"source_id":3946,"source_name":3936,"source_type":91,"source_url":3947,"stem":3948,"tags":3949,"thumbnail_url":56,"tldr":3950,"tweet":56,"unknown_tags":3951,"__hash__":3952},"summaries\u002Fsummaries\u002F9cf4eabf30c8f73e-open-source-ai-innovation-engine-or-security-risk-summary.md","Open Source AI: Innovation Engine or Security Risk?",{"provider":7,"model":8,"input_tokens":3818,"output_tokens":3819,"processing_time_ms":3820,"cost_usd":3821},8403,2068,28390,0.00242375,{"type":14,"value":3823,"toc":3919},[3824,3828,3831,3834,3838,3841,3844,3847,3851,3854,3857,3860,3864,3867,3870,3873,3892,3896],[17,3825,3827],{"id":3826},"open-source-as-ais-innovation-backbone","Open Source as AI's Innovation Backbone",[22,3829,3830],{},"Gabe Goodhart champions open source as the core of AI progress, arguing that AI's science—math and tensors—is unusually close to usable products, unlike past tech waves. This tight coupling means open source hosts most real innovation, with examples like linear attention mechanisms enabling better long-context models through collaborative tweaks (e.g., hybridizing recurrent linear layers with attention). Martin Keen reinforces this, noting open source underpins even closed frontier models via standards like Anthropic's Model Context Protocol (MCP), donated to Linux Foundation, and agent skill specs like skill.md. These allow open-weight models (Mistral, Llama, Deepseek) or closed ones to invoke services uniformly. All panelists concur: open source democratizes access, accelerates catch-up to frontier capabilities, and fosters architectures anyone can run on their hardware, bypassing lab gatekeeping.",[22,3832,3833],{},"Jeff Crume, the security skeptic, doesn't dispute innovation benefits but tempers hype. He invokes Kerckhoffs' principle from cryptography: only keys should be secret, not algorithms, as scrutiny strengthens systems. Yet he cautions Linux's history—once claimed malware-proof—shows open source isn't secure by default. Consensus emerges: open source thrives on 'a thousand eyes' for innovation, but scale (billions of parameters) overwhelms full vetting.",[17,3835,3837],{"id":3836},"secure-vs-securable-core-security-distinction","Secure vs. Securable: Core Security Distinction",[22,3839,3840],{},"Jeff Crume draws a sharp line: open source AI is 'securable' (design allows fixes) but not 'secure' without deliberate controls. Proprietary claims of inherent security fare no better; security stems from implementation, not source status. Transparency builds trust, essential for AI, but latent bugs in decades-old open code prove even crowds miss flaws. AI itself aids detection—scanning source or reverse-engineering binaries via LLMs\u002Fdecompilers, a capability predating gen AI but now amplified.",[22,3842,3843],{},"Gabe echoes: AI stacks mirror Linux\u002FKubernetes—composable open projects with attack surfaces needing updates and policies. Open code enables fixes, but poor projects emit 'vibe code Spidey sense' (unmanaged security). Martin highlights hybrid realities: closed models rely on open foundations, blurring lines. Divergence: Gabe sees open weights accelerating science (e.g., attention innovations), while Jeff notes proprietary guardrails (pre\u002Fpost-model filters) block misuse—open weights invite 'obliteration' of safety layers via embedding tweaks.",[22,3845,3846],{},"Panel agrees on trust: opacity breeds blind faith, not security. Open source invites scrutiny, but demands proactive policy.",[17,3848,3850],{"id":3849},"model-access-bad-actors-and-emerging-threats","Model Access, Bad Actors, and Emerging Threats",[22,3852,3853],{},"Debate heats on access: frontier models (e.g., latest from labs) gatekeep via approvals\u002Fconsortia, while open models on Hugging Face run anywhere. Martin predicts open source will close gaps quickly. Jeff worries bad actors access simultaneously—security through obscurity fails, as leaks inevitable. Yet open weights expose more: attackers strip refusals, unleashing unfiltered capabilities.",[22,3855,3856],{},"Gabe ties to agents: autonomy turns agent loops into code interpreters where 'the internet is your untrusted code.' Textual inputs, once filtered by humans\u002Fprograms, now trigger actions via tools—massive attack surface. OpenClaw-like systems exemplify chaos. Jeff nods to AI's dual role: vulnerability scanner and exploit amplifier (reverse-engineering binaries). Martin\u002F Gabe stress context layers (beyond models\u002Fsoftware) compound risks.",[22,3858,3859],{},"Strongest arguments: Pro-open (Gabe\u002FMartin)—stifling access hampers progress; security (Jeff)—scale defeats 'many eyes,' demands controls. No one-size-fits-all; mitigate via guardrails, sandboxes, updates.",[17,3861,3863],{"id":3862},"using-ai-to-secure-ai-and-forward-outlook","Using AI to Secure AI and Forward Outlook",[22,3865,3866],{},"Jeff sees AI securing itself as nuanced: LLMs find code vulns faster than humans, even in proprietary binaries (decompile → scan). Not new—pre-gen AI tools existed—but gen AI scales it. Gabe warns of net-new agent risks, urging trust-boundary rethinking.",[22,3868,3869],{},"Predictions: Open models catch frontiers; innovation via open science\u002Farchitectures unstoppable. Recommendations: Vet projects rigorously; sandbox agents; stay updated; blend open\u002Fclosed (e.g., open standards with closed models). Tradeoffs: Open weights boost utility\u002Finnovation but heighten misuse; closed offers controls at velocity cost.",[22,3871,3872],{},"Notable quotes:",[3874,3875,3876,3880,3883,3886,3889],"ul",{},[3877,3878,3879],"li",{},"Gabe Goodhart: \"Open-source relative to most other innovation waves is where the vast majority of the actual innovation is happening because science by its very nature is open.\"",[3877,3881,3882],{},"Jeff Crume: \"Linux is a good example of a system that is securable, but in and of itself is not necessarily secure.\"",[3877,3884,3885],{},"Gabe Goodhart: \"The agent loop is essentially a code interpreter and the code is literally any text you pass through it... now the internet is your untrusted code.\"",[3877,3887,3888],{},"Jeff Crume: \"Security through obscurity is not an effective model... the only thing about a crypto system that should be secret are the keys.\"",[3877,3890,3891],{},"Martin Keen: \"Open source is foundational to everything in AI now even if we're talking about models that were actually frontier closed models.\"",[17,3893,3895],{"id":3894},"key-takeaways","Key Takeaways",[3874,3897,3898,3901,3904,3907,3910,3913,3916],{},[3877,3899,3900],{},"Prioritize 'securable' open source projects with strong security contribution policies and update cadences—avoid vibe-check fails.",[3877,3902,3903],{},"Distinguish open code (innovation accelerator) from open weights (misuse risk)—use guardrails for models, sandboxes for agents.",[3877,3905,3906],{},"Leverage AI for vuln scanning on open or closed code; reverse-engineering erodes proprietary edges.",[3877,3908,3909],{},"Blend approaches: Open standards (MCP, skill.md) enhance closed models; run open weights on your infra for flexibility.",[3877,3911,3912],{},"Build trust via transparency and controls, not secrecy—bad actors leak anyway; focus on implementation.",[3877,3914,3915],{},"For agents, treat all inputs as untrusted code—rethink textual data assumptions.",[3877,3917,3918],{},"Expect open models to trail but catch frontier capabilities quickly via collaborative innovation.",{"title":47,"searchDepth":48,"depth":48,"links":3920},[3921,3922,3923,3924,3925],{"id":3826,"depth":48,"text":3827},{"id":3836,"depth":48,"text":3837},{"id":3849,"depth":48,"text":3850},{"id":3862,"depth":48,"text":3863},{"id":3894,"depth":48,"text":3895},[55],{"content_references":3928,"triage":3938},[3929,3934],{"type":3930,"title":3931,"author":3932,"url":3933,"context":65},"podcast","Security Intelligence x Mixture of Experts Crossover Episode","IBM Technology (Matt Kazinski host)","https:\u002F\u002Fibm.biz\u002F~sTfk9xICA",{"type":3930,"title":3935,"author":3936,"url":3937,"context":3798},"Mixture of Experts","IBM Technology","https:\u002F\u002Fibm.biz\u002F~SMOMF0sqx",{"relevance":3739,"novelty":3739,"quality":81,"actionability":48,"composite":3939,"reasoning":3940},3.05,"Category: AI & LLMs. The article discusses the role of open source in AI innovation and security, which aligns with the AI & LLMs category. While it presents some new perspectives on the security risks associated with open source AI, it lacks specific actionable steps for the audience to implement in their projects.","\u002Fsummaries\u002F9cf4eabf30c8f73e-open-source-ai-innovation-engine-or-security-risk-summary","2026-04-29 10:00:42","2026-05-03 16:43:49",{"title":3816,"description":47},{"loc":3941},"e15ac1bd93fe2101","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=PJGIWDW_W2A","summaries\u002F9cf4eabf30c8f73e-open-source-ai-innovation-engine-or-security-risk-summary",[98,95,97],"Panelists agree open source drives AI breakthroughs but warn it's 'securable' not 'secure'—needs rigorous practices to mitigate risks like model tampering and agent exploits.",[],"k2Bl5Kceueq5fD2_72bTo7FXkyKIWmwztjNZnjDyot8",{"id":3954,"title":3955,"ai":3956,"body":3961,"categories":4008,"created_at":56,"date_modified":56,"description":47,"extension":57,"faq":56,"featured":58,"kicker_label":56,"meta":4009,"navigation":84,"path":4017,"published_at":4018,"question":56,"scraped_at":4019,"seo":4020,"sitemap":4021,"source_id":4022,"source_name":4023,"source_type":91,"source_url":4024,"stem":4025,"tags":4026,"thumbnail_url":56,"tldr":4027,"tweet":56,"unknown_tags":4028,"__hash__":4029},"summaries\u002Fsummaries\u002Ffbbbc098d7e53ea7-hermes-agent-persists-learning-across-sessions-summary.md","Hermes Agent Persists Learning Across Sessions",{"provider":7,"model":8,"input_tokens":3957,"output_tokens":3958,"processing_time_ms":3959,"cost_usd":3960},3887,1356,12936,0.0014329,{"type":14,"value":3962,"toc":4003},[3963,3967,3970,3974,3982,3993,3996,4000],[17,3964,3966],{"id":3965},"session-amnesia-limits-current-ai-agents","Session Amnesia Limits Current AI Agents",[22,3968,3969],{},"Most AI agents today erase user-specific knowledge—like your tech stack, naming conventions, server details, or preferences—after each session. This forces repetitive context pasting, starting every conversation from scratch and wasting time on rediscovering basics. The result: agents feel like strangers, unable to build on prior help despite nodding along during interactions.",[17,3971,3973],{"id":3972},"hermes-builds-reusable-knowledge-via-learning-loop","Hermes Builds Reusable Knowledge via Learning Loop",[22,3975,3976,3977,3981],{},"Hermes Agent, an open-source project by Nous Research, embeds persistence from the ground up. Its core mechanism is a ",[3978,3979,3980],"em",{},"learning loop"," that:",[3874,3983,3984,3987,3990],{},[3877,3985,3986],{},"Records what worked in interactions.",[3877,3988,3989],{},"Distills those into reusable procedures.",[3877,3991,3992],{},"Automatically loads relevant procedures for matching future problems.",[22,3994,3995],{},"This isn't a bolted-on memory feature but a foundational design, turning one-off help into scalable, context-aware automation. Builders get an agent that evolves with use, reducing setup friction over time.",[17,3997,3999],{"id":3998},"practical-value-for-ai-agent-builders","Practical Value for AI Agent Builders",[22,4001,4002],{},"Hermes stands out among agents by addressing real-world retention gaps, making it ideal for ongoing workflows like coding or ops. Exploring its mechanics reveals patterns for your own agents: prioritize procedure extraction over raw chat history to enable true adaptation. If shipping persistent AI tools, benchmark against Hermes to avoid common forgetfulness pitfalls—it's a concrete step toward agents that compound value across sessions.",{"title":47,"searchDepth":48,"depth":48,"links":4004},[4005,4006,4007],{"id":3965,"depth":48,"text":3966},{"id":3972,"depth":48,"text":3973},{"id":3998,"depth":48,"text":3999},[55],{"content_references":4010,"triage":4015},[4011],{"type":3728,"title":4012,"author":4013,"url":4014,"context":65},"Hermes Agent","Nous Research","https:\u002F\u002Fnousresearch.com\u002F",{"relevance":80,"novelty":81,"quality":81,"actionability":81,"composite":82,"reasoning":4016},"Category: AI & LLMs. The article discusses Hermes, an AI agent that learns across sessions, addressing a significant pain point for developers building AI tools. It provides actionable insights on how to implement a learning loop in AI agents, making it highly relevant and practical for the target audience.","\u002Fsummaries\u002Ffbbbc098d7e53ea7-hermes-agent-persists-learning-across-sessions-summary","2026-04-21 14:01:02","2026-04-21 15:26:08",{"title":3955,"description":47},{"loc":4017},"fbbbc098d7e53ea7","Towards AI","https:\u002F\u002Fpub.towardsai.net\u002Fthe-ai-agent-that-actually-learns-from-you-inside-hermes-by-nous-research-fd074717a8e7?source=rss----98111c9905da---4","summaries\u002Ffbbbc098d7e53ea7-hermes-agent-persists-learning-across-sessions-summary",[97,95,98],"Unlike typical AI agents that reset context per session, Hermes from Nous Research uses a learning loop to capture successful procedures from interactions and auto-apply them to similar future tasks.",[],"lsDq90yZPxLIKlzGTbUAmJbeqmIwISt1vnbqR2XZ_TQ"]