[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"summary-2f3ae40d3bb79a21-gemma-4-efficient-multimodal-open-llms-for-edge-to-summary":3,"summaries-facets-categories":183,"summary-related-2f3ae40d3bb79a21-gemma-4-efficient-multimodal-open-llms-for-edge-to-summary":3753},{"id":4,"title":5,"ai":6,"body":13,"categories":131,"created_at":132,"date_modified":132,"description":125,"extension":133,"faq":132,"featured":134,"kicker_label":132,"meta":135,"navigation":166,"path":167,"published_at":132,"question":132,"scraped_at":168,"seo":169,"sitemap":170,"source_id":171,"source_name":172,"source_type":173,"source_url":174,"stem":175,"tags":176,"thumbnail_url":132,"tldr":180,"tweet":132,"unknown_tags":181,"__hash__":182},"summaries\u002Fsummaries\u002F2f3ae40d3bb79a21-gemma-4-efficient-multimodal-open-llms-for-edge-to-summary.md","Gemma 4: Efficient Multimodal Open LLMs for Edge to Server",{"provider":7,"model":8,"input_tokens":9,"output_tokens":10,"processing_time_ms":11,"cost_usd":12},"openrouter","x-ai\u002Fgrok-4.1-fast",9628,1838,16028,0.0028221,{"type":14,"value":15,"toc":124},"minimark",[16,21,25,29,32,36,39,121],[17,18,20],"h2",{"id":19},"tailored-architectures-balance-capability-and-deployment","Tailored Architectures Balance Capability and Deployment",[22,23,24],"p",{},"Gemma 4 spans three architectures to match hardware constraints: E2B and E4B effective-parameter models (using Per-Layer Embeddings for token-efficient lookups) target ultra-mobile edge like Pixel or Chrome browsers; 31B dense model enables server-grade performance locally; 26B A4B Mixture-of-Experts activates only 4B parameters per token but loads all 26B for fast routing. Download weights from Kaggle or Hugging Face for commercial use with tuning. Prioritize E2B\u002FE4B for on-device to cut memory via PLE, which inflates static weights but speeds lookups over deeper layers.",[17,26,28],{"id":27},"multimodal-reasoning-with-extended-context","Multimodal Reasoning with Extended Context",[22,30,31],{},"All models excel in reasoning via configurable thinking modes, coding benchmarks, and native function calling for agents—generate structured tool calls directly. Input text universally; images (variable aspect\u002Fresolution) across all; video and audio native to E2B\u002FE4B. Context reaches 128K tokens on small models, 256K on medium for long conversations or documents. Use built-in system prompts for controlled chats, outperforming prior Gemma via native support over legacy formatting.",[17,33,35],{"id":34},"quantization-drives-memory-trade-offs","Quantization Drives Memory Trade-offs",[22,37,38],{},"Run at BF16 default or quantize to SFP8\u002FQ4_0 for efficiency—higher bits boost accuracy but spike compute\u002Fmemory\u002Fpower. Base weights exclude KV cache (scales with prompt\u002Fresponse tokens) or fine-tuning overhead (use LoRA\u002FPEFT to minimize). MoE loads full params despite sparse activation.",[40,41,42,61],"table",{},[43,44,45],"thead",{},[46,47,48,52,55,58],"tr",{},[49,50,51],"th",{},"Parameters",[49,53,54],{},"BF16 (16-bit)",[49,56,57],{},"SFP8 (8-bit)",[49,59,60],{},"Q4_0 (4-bit)",[62,63,64,79,93,107],"tbody",{},[46,65,66,70,73,76],{},[67,68,69],"td",{},"Gemma 4 E2B",[67,71,72],{},"9.6 GB",[67,74,75],{},"4.6 GB",[67,77,78],{},"3.2 GB",[46,80,81,84,87,90],{},[67,82,83],{},"Gemma 4 E4B",[67,85,86],{},"15 GB",[67,88,89],{},"7.5 GB",[67,91,92],{},"5 GB",[46,94,95,98,101,104],{},[67,96,97],{},"Gemma 4 31B",[67,99,100],{},"58.3 GB",[67,102,103],{},"30.4 GB",[67,105,106],{},"17.4 GB",[46,108,109,112,115,118],{},[67,110,111],{},"Gemma 4 26B A4B",[67,113,114],{},"48 GB",[67,116,117],{},"25 GB",[67,119,120],{},"15.6 GB",[22,122,123],{},"Plan VRAM as base + context; test in your stack since tools vary. Previous Gemma 1-3 available for legacy needs.",{"title":125,"searchDepth":126,"depth":126,"links":127},"",2,[128,129,130],{"id":19,"depth":126,"text":20},{"id":27,"depth":126,"text":28},{"id":34,"depth":126,"text":35},[],null,"md",false,{"content_references":136,"triage":160},[137,142,146,149,152,157],{"type":138,"title":139,"url":140,"context":141},"other","Gemma 4 Model Card","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore\u002Fmodel_card_4","cited",{"type":138,"title":143,"url":144,"context":145},"Gemma 3 Model Card","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore\u002Fmodel_card_3","mentioned",{"type":138,"title":147,"url":148,"context":145},"Gemma 2 Model Card","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore\u002Fmodel_card_2",{"type":138,"title":150,"url":151,"context":145},"Gemma 1 Model Card","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore\u002Fmodel_card",{"type":153,"title":154,"url":155,"context":156},"tool","Kaggle","https:\u002F\u002Fwww.kaggle.com\u002Fmodels?query=gemma-4&publisher=google","recommended",{"type":153,"title":158,"url":159,"context":156},"Hugging Face","https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fgoogle\u002Fgemma-4",{"relevance":161,"novelty":162,"quality":162,"actionability":163,"composite":164,"reasoning":165},5,4,3,4.15,"Category: AI & LLMs. The article provides in-depth information about the Gemma 4 models, detailing their architectures and capabilities, which directly addresses the needs of developers looking to integrate AI into their products. It includes specific technical details such as memory requirements and quantization options, making it actionable, though it lacks step-by-step guidance for implementation.",true,"\u002Fsummaries\u002F2f3ae40d3bb79a21-gemma-4-efficient-multimodal-open-llms-for-edge-to-summary","2026-04-15 15:33:13",{"title":5,"description":125},{"loc":167},"2f3ae40d3bb79a21","__oneoff__","article","https:\u002F\u002Fai.google.dev\u002Fgemma\u002Fdocs\u002Fcore","summaries\u002F2f3ae40d3bb79a21-gemma-4-efficient-multimodal-open-llms-for-edge-to-summary",[177,178,179],"llm","agents","open-source","Gemma 4 delivers open-weight models in 2B\u002F4B effective (edge-optimized), 31B dense, and 26B MoE sizes with text\u002Fimage\u002Fvideo\u002Faudio input, 128K-256K context, function calling, and quantization down to 3.2GB memory for E2B inference.",[],"V1eLeg2WWIShcBjn4EKeTTR1xM0puKG1JLMTVe9OxzI",[184,187,190,193,196,199,201,203,205,207,209,211,214,216,218,220,222,224,226,228,230,232,235,238,240,242,245,247,249,252,254,256,258,260,262,264,266,268,270,272,274,276,278,280,282,284,286,288,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338,340,342,344,346,348,350,352,354,356,358,360,362,364,366,368,370,372,374,376,378,380,382,384,386,388,390,392,394,396,398,400,402,404,406,408,410,412,414,416,418,420,422,424,426,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,458,460,462,464,466,468,470,472,474,476,478,480,482,484,486,488,490,492,494,496,498,500,502,504,507,509,511,513,515,517,519,521,523,525,527,529,531,533,535,537,539,541,543,545,547,549,551,553,555,557,559,561,563,565,567,569,571,573,575,577,579,581,583,585,587,589,591,593,595,597,599,601,603,605,607,609,611,613,615,617,619,621,623,625,627,629,631,633,635,637,639,641,643,645,647,649,651,653,655,657,659,661,663,665,667,669,671,673,675,677,679,681,683,685,687,689,691,693,695,697,699,701,703,705,707,709,711,713,715,717,719,721,723,725,727,729,731,733,735,737,739,741,743,745,747,749,751,753,755,757,759,761,763,765,767,769,771,773,775,777,779,781,783,785,787,789,791,793,795,797,799,801,803,805,807,809,811,813,815,817,819,821,823,825,827,829,831,833,835,837,839,841,843,845,847,849,851,853,855,857,859,861,863,865,867,869,871,873,875,877,879,881,883,885,887,889,891,893,895,897,899,901,903,905,907,909,911,913,915,917,919,921,923,925,927,929,931,933,935,937,939,941,943,945,947,949,951,953,955,957,959,961,963,965,967,969,971,973,975,977,979,981,983,985,987,989,991,993,995,997,999,1001,1003,1005,1007,1009,1011,1013,1015,1017,1019,1021,1023,1025,1027,1029,1031,1033,1035,1037,1039,1041,1043,1045,1047,1049,1051,1053,1055,1057,1059,1061,1063,1065,1067,1069,1071,1073,1075,1077,1079,1081,1083,1085,1087,1089,1091,1093,1095,1097,1099,1101,1103,1105,1107,1109,1111,1113,1115,1117,1119,1121,1123,1125,1127,1129,1131,1133,1135,1137,1139,1141,1143,1145,1147,1149,1151,1153,1155,1157,1159,1161,1163,1165,1167,1169,1171,1173,1175,1177,1179,1181,1183,1185,1187,1189,1191,1193,1195,1197,1199,1201,1203,1205,1207,1209,1211,1213,1215,1217,1219,1221,1223,1225,1227,1229,1231,1233,1235,1237,1239,1241,1243,1245,1247,1249,1251,1253,1255,1257,1259,1261,1263,1265,1267,1269,1271,1273,1275,1277,1279,1281,1283,1285,1287,1289,1291,1293,1295,1297,1299,1301,1303,1305,1307,1309,1311,1313,1315,1317,1319,1321,1323,1325,1327,1329,1331,1333,1335,1337,1339,1341,1343,1345,1347,1349,1351,1353,1355,1357,1359,1361,1363,1365,1367,1369,1371,1373,1375,1377,1379,1381,1383,1385,1387,1389,1391,1393,1395,1397,1399,1401,1403,1405,1407,1409,1411,1413,1415,1417,1419,1421,1423,1425,1427,1429,1431,1433,1435,1437,1439,1441,1443,1445,1447,1449,1451,1453,1455,1457,1459,1461,1463,1465,1467,1469,1471,1473,1475,1477,1479,1481,1483,1485,1487,1489,1491,1493,1495,1497,1499,1501,1503,1505,1507,1509,1511,1513,1515,1517,1519,1521,1523,1525,1527,1529,1531,1533,1535,1537,1539,1541,1543,1545,1547,1549,1551,1553,1555,1557,1559,1561,1563,1565,1567,1569,1571,1573,1575,1577,1579,1581,1583,1585,1587,1589,1591,1593,1595,1597,1599,1601,1603,1605,1607,1609,1611,1613,1615,1617,1619,1621,1623,1625,1627,1629,1631,1633,1635,1637,1639,1641,1643,1645,1647,1649,1651,1653,1655,1657,1659,1661,1663,1665,1667,1669,1671,1673,1675,1677,1679,1681,1683,1685,1687,1689,1691,1693,1695,1697,1699,1701,1703,1705,1707,1709,1711,1713,1715,1717,1719,1721,1723,1725,1727,1729,1731,1733,1735,1737,1739,1741,1743,1745,1747,1749,1751,1753,1755,1757,1759,1761,1763,1765,1767,1769,1771,1773,1775,1777,1779,1781,1783,1785,1787,1789,1791,1793,1795,1797,1799,1801,1803,1805,1807,1809,1811,1813,1815,1817,1819,1821,1823,1825,1827,1829,1831,1833,1835,1837,1839,1841,1843,1845,1847,1849,1851,1853,1855,1857,1859,1861,1863,1865,1867,1869,1871,1873,1875,1877,1879,1881,1883,1885,1887,1889,1891,1893,1895,1897,1899,1901,1903,1905,1907,1909,1911,1913,1915,1917,1919,1921,1923,1925,1927,1929,1931,1933,1935,1937,1939,1941,1943,1945,1947,1949,1951,1953,1955,1957,1959,1961,1963,1965,1967,1969,1971,1973,1975,1977,1979,1981,1983,1985,1987,1989,1991,1993,1995,1997,1999,2001,2003,2005,2007,2009,2011,2013,2015,2017,2019,2021,2023,2025,2027,2029,2031,2033,2035,2037,2039,2041,2043,2045,2047,2049,2051,2053,2055,2057,2059,2061,2063,2065,2067,2069,2071,2073,2075,2077,2079,2081,2083,2085,2087,2089,2091,2093,2095,2097,2099,2101,2103,2105,2107,2109,2111,2113,2115,2117,2119,2121,2123,2125,2127,2129,2131,2133,2135,2137,2139,2141,2143,2145,2147,2149,2151,2153,2155,2157,2159,2161,2163,2165,2167,2169,2171,2173,2175,2177,2179,2181,2183,2185,2187,2189,2191,2193,2195,2197,2199,2201,2203,2205,2207,2209,2211,2213,2215,2217,2219,2221,2223,2225,2227,2229,2231,2233,2235,2237,2239,2241,2243,2245,2247,2249,2251,2253,2255,2257,2259,2261,2263,2265,2267,2269,2271,2273,2275,2277,2279,2281,2283,2285,2287,2289,2291,2293,2295,2297,2299,2301,2303,2305,2307,2309,2311,2313,2315,2317,2319,2321,2323,2325,2327,2329,2331,2333,2335,2337,2339,2341,2343,2345,2347,2349,2351,2353,2355,2357,2359,2361,2363,2365,2367,2369,2371,2373,2375,2377,2379,2381,2383,2385,2387,2389,2391,2393,2395,2397,2399,2401,2403,2405,2407,2409,2411,2413,2415,2417,2419,2421,2423,2425,2427,2429,2431,2433,2435,2437,2439,2441,2443,2445,2447,2449,2451,2453,2455,2457,2459,2461,2463,2465,2467,2469,2471,2473,2475,2477,2479,2481,2483,2485,2487,2489,2491,2493,2495,2497,2499,2501,2503,2505,2507,2509,2511,2513,2515,2517,2519,2521,2523,2525,2527,2529,2531,2533,2535,2537,2539,2541,2543,2545,2547,2549,2551,2553,2555,2557,2559,2561,2563,2565,2567,2569,2571,2573,2575,2577,2579,2581,2583,2585,2587,2589,2591,2593,2595,2597,2599,2601,2603,2605,2607,2609,2611,2613,2615,2617,2619,2621,2623,2625,2627,2629,2631,2633,2635,2637,2639,2641,2643,2645,2647,2649,2651,2653,2655,2657,2659,2661,2663,2665,2667,2669,2671,2673,2675,2677,2679,2681,2683,2685,2687,2689,2691,2693,2695,2697,2699,2701,2703,2705,2707,2709,2711,2713,2715,2717,2719,2721,2723,2725,2727,2729,2731,2733,2735,2737,2739,2741,2743,2745,2747,2749,2751,2753,2755,2757,2759,2761,2763,2765,2767,2769,2771,2773,2775,2777,2779,2781,2783,2785,2787,2789,2791,2793,2795,2797,2799,2801,2803,2805,2807,2809,2811,2813,2815,2817,2819,2821,2823,2825,2827,2829,2831,2833,2835,2837,2839,2841,2843,2845,2847,2849,2851,2853,2855,2857,2859,2861,2863,2865,2867,2869,2871,2873,2875,2877,2879,2881,2883,2885,2887,2889,2891,2893,2895,2897,2899,2901,2903,2905,2907,2909,2911,2913,2915,2917,2919,2921,2923,2925,2927,2929,2931,2933,2935,2937,2939,2941,2943,2945,2947,2949,2951,2953,2955,2957,2959,2961,2963,2965,2967,2969,2971,2973,2975,2977,2979,2981,2983,2985,2987,2989,2991,2993,2995,2997,2999,3001,3003,3005,3007,3009,3011,3013,3015,3017,3019,3021,3023,3025,3027,3029,3031,3033,3035,3037,3039,3041,3043,3045,3047,3049,3051,3053,3055,3057,3059,3061,3063,3065,3067,3069,3071,3073,3075,3077,3079,3081,3083,3085,3087,3089,3091,3093,3095,3097,3099,3101,3103,3105,3107,3109,3111,3113,3115,3117,3119,3121,3123,3125,3127,3129,3131,3133,3135,3137,3139,3141,3143,3145,3147,3149,3151,3153,3155,3157,3159,3161,3163,3165,3167,3169,3171,3173,3175,3177,3179,3181,3183,3185,3187,3189,3191,3193,3195,3197,3199,3201,3203,3205,3207,3209,3211,3213,3215,3217,3219,3221,3223,3225,3227,3229,3231,3233,3235,3237,3239,3241,3243,3245,3247,3249,3251,3253,3255,3257,3259,3261,3263,3265,3267,3269,3271,3273,3275,3277,3279,3281,3283,3285,3287,3289,3291,3293,3295,3297,3299,3301,3303,3305,3307,3309,3311,3313,3315,3317,3319,3321,3323,3325,3327,3329,3331,3333,3335,3337,3339,3341,3343,3345,3347,3349,3351,3353,3355,3357,3359,3361,3363,3365,3367,3369,3371,3373,3375,3377,3379,3381,3383,3385,3387,3389,3391,3393,3395,3397,3399,3401,3403,3405,3407,3409,3411,3413,3415,3417,3419,3421,3423,3425,3427,3429,3431,3433,3435,3437,3439,3441,3443,3445,3447,3449,3451,3453,3455,3457,3459,3461,3463,3465,3467,3469,3471,3473,3475,3477,3479,3481,3483,3485,3487,3489,3491,3493,3495,3497,3499,3501,3503,3505,3507,3509,3511,3513,3515,3517,3519,3521,3523,3525,3527,3529,3531,3533,3535,3537,3539,3541,3543,3545,3547,3549,3551,3553,3555,3557,3559,3561,3563,3565,3567,3569,3571,3573,3575,3577,3579,3581,3583,3585,3587,3589,3591,3593,3595,3597,3599,3601,3603,3605,3607,3609,3611,3613,3615,3617,3619,3621,3623,3625,3627,3629,3631,3633,3635,3637,3639,3641,3643,3645,3647,3649,3651,3653,3655,3657,3659,3661,3663,3665,3667,3669,3671,3673,3675,3677,3679,3681,3683,3685,3687,3689,3691,3693,3695,3697,3699,3701,3703,3705,3707,3709,3711,3713,3715,3717,3719,3721,3723,3725,3727,3729,3731,3733,3735,3737,3739,3741,3743,3745,3747,3749,3751],{"categories":185},[186],"Developer Productivity",{"categories":188},[189],"Business & SaaS",{"categories":191},[192],"AI & LLMs",{"categories":194},[195],"AI Automation",{"categories":197},[198],"Product Strategy",{"categories":200},[192],{"categories":202},[186],{"categories":204},[189],{"categories":206},[],{"categories":208},[192],{"categories":210},[],{"categories":212},[213],"AI News & Trends",{"categories":215},[195],{"categories":217},[213],{"categories":219},[195],{"categories":221},[195],{"categories":223},[192],{"categories":225},[192],{"categories":227},[213],{"categories":229},[192],{"categories":231},[],{"categories":233},[234],"Design & Frontend",{"categories":236},[237],"Data Science & Visualization",{"categories":239},[213],{"categories":241},[],{"categories":243},[244],"Software Engineering",{"categories":246},[192],{"categories":248},[195],{"categories":250},[251],"Marketing & Growth",{"categories":253},[192],{"categories":255},[195],{"categories":257},[],{"categories":259},[],{"categories":261},[234],{"categories":263},[195],{"categories":265},[186],{"categories":267},[234],{"categories":269},[192],{"categories":271},[195],{"categories":273},[213],{"categories":275},[],{"categories":277},[],{"categories":279},[195],{"categories":281},[244],{"categories":283},[],{"categories":285},[189],{"categories":287},[],{"categories":289},[],{"categories":291},[195],{"categories":293},[195],{"categories":295},[192],{"categories":297},[],{"categories":299},[244],{"categories":301},[],{"categories":303},[],{"categories":305},[],{"categories":307},[192],{"categories":309},[251],{"categories":311},[234],{"categories":313},[234],{"categories":315},[192],{"categories":317},[195],{"categories":319},[192],{"categories":321},[192],{"categories":323},[195],{"categories":325},[195],{"categories":327},[237],{"categories":329},[213],{"categories":331},[195],{"categories":333},[251],{"categories":335},[195],{"categories":337},[198],{"categories":339},[],{"categories":341},[195],{"categories":343},[],{"categories":345},[195],{"categories":347},[244],{"categories":349},[234],{"categories":351},[192],{"categories":353},[],{"categories":355},[],{"categories":357},[195],{"categories":359},[],{"categories":361},[192],{"categories":363},[],{"categories":365},[186],{"categories":367},[244],{"categories":369},[189],{"categories":371},[213],{"categories":373},[192],{"categories":375},[],{"categories":377},[192],{"categories":379},[],{"categories":381},[244],{"categories":383},[237],{"categories":385},[],{"categories":387},[192],{"categories":389},[234],{"categories":391},[],{"categories":393},[234],{"categories":395},[195],{"categories":397},[],{"categories":399},[195],{"categories":401},[213],{"categories":403},[192],{"categories":405},[],{"categories":407},[195],{"categories":409},[192],{"categories":411},[198],{"categories":413},[],{"categories":415},[192],{"categories":417},[195],{"categories":419},[195],{"categories":421},[],{"categories":423},[237],{"categories":425},[192],{"categories":427},[],{"categories":429},[186],{"categories":431},[189],{"categories":433},[192],{"categories":435},[195],{"categories":437},[244],{"categories":439},[192],{"categories":441},[],{"categories":443},[],{"categories":445},[192],{"categories":447},[],{"categories":449},[234],{"categories":451},[],{"categories":453},[192],{"categories":455},[],{"categories":457},[195],{"categories":459},[192],{"categories":461},[234],{"categories":463},[],{"categories":465},[192],{"categories":467},[192],{"categories":469},[189],{"categories":471},[195],{"categories":473},[192],{"categories":475},[234],{"categories":477},[195],{"categories":479},[],{"categories":481},[],{"categories":483},[213],{"categories":485},[],{"categories":487},[192],{"categories":489},[189,251],{"categories":491},[],{"categories":493},[192],{"categories":495},[],{"categories":497},[],{"categories":499},[192],{"categories":501},[],{"categories":503},[192],{"categories":505},[506],"DevOps & Cloud",{"categories":508},[],{"categories":510},[213],{"categories":512},[234],{"categories":514},[],{"categories":516},[213],{"categories":518},[213],{"categories":520},[192],{"categories":522},[251],{"categories":524},[],{"categories":526},[189],{"categories":528},[],{"categories":530},[192,506],{"categories":532},[192],{"categories":534},[192],{"categories":536},[195],{"categories":538},[192,244],{"categories":540},[237],{"categories":542},[192],{"categories":544},[251],{"categories":546},[195],{"categories":548},[195],{"categories":550},[],{"categories":552},[195],{"categories":554},[192,189],{"categories":556},[],{"categories":558},[234],{"categories":560},[234],{"categories":562},[],{"categories":564},[],{"categories":566},[213],{"categories":568},[],{"categories":570},[186],{"categories":572},[244],{"categories":574},[192],{"categories":576},[234],{"categories":578},[195],{"categories":580},[244],{"categories":582},[213],{"categories":584},[234],{"categories":586},[],{"categories":588},[192],{"categories":590},[192],{"categories":592},[192],{"categories":594},[213],{"categories":596},[186],{"categories":598},[192],{"categories":600},[195],{"categories":602},[506],{"categories":604},[234],{"categories":606},[195],{"categories":608},[],{"categories":610},[],{"categories":612},[234],{"categories":614},[213],{"categories":616},[237],{"categories":618},[],{"categories":620},[192],{"categories":622},[192],{"categories":624},[189],{"categories":626},[192],{"categories":628},[192],{"categories":630},[213],{"categories":632},[],{"categories":634},[195],{"categories":636},[244],{"categories":638},[],{"categories":640},[192],{"categories":642},[192],{"categories":644},[195],{"categories":646},[],{"categories":648},[],{"categories":650},[192],{"categories":652},[],{"categories":654},[189],{"categories":656},[195],{"categories":658},[],{"categories":660},[186],{"categories":662},[192],{"categories":664},[189],{"categories":666},[213],{"categories":668},[],{"categories":670},[],{"categories":672},[],{"categories":674},[213],{"categories":676},[213],{"categories":678},[],{"categories":680},[],{"categories":682},[189],{"categories":684},[],{"categories":686},[],{"categories":688},[186],{"categories":690},[],{"categories":692},[251],{"categories":694},[195],{"categories":696},[189],{"categories":698},[195],{"categories":700},[],{"categories":702},[198],{"categories":704},[234],{"categories":706},[244],{"categories":708},[192],{"categories":710},[195],{"categories":712},[189],{"categories":714},[192],{"categories":716},[],{"categories":718},[],{"categories":720},[244],{"categories":722},[237],{"categories":724},[198],{"categories":726},[195],{"categories":728},[192],{"categories":730},[],{"categories":732},[506],{"categories":734},[],{"categories":736},[195],{"categories":738},[],{"categories":740},[],{"categories":742},[192],{"categories":744},[234],{"categories":746},[251],{"categories":748},[195],{"categories":750},[],{"categories":752},[186],{"categories":754},[],{"categories":756},[213],{"categories":758},[192,506],{"categories":760},[213],{"categories":762},[192],{"categories":764},[189],{"categories":766},[192],{"categories":768},[],{"categories":770},[189],{"categories":772},[],{"categories":774},[244],{"categories":776},[234],{"categories":778},[213],{"categories":780},[237],{"categories":782},[186],{"categories":784},[192],{"categories":786},[244],{"categories":788},[],{"categories":790},[],{"categories":792},[198],{"categories":794},[],{"categories":796},[192],{"categories":798},[],{"categories":800},[234],{"categories":802},[234],{"categories":804},[234],{"categories":806},[],{"categories":808},[],{"categories":810},[213],{"categories":812},[195],{"categories":814},[192],{"categories":816},[192],{"categories":818},[192],{"categories":820},[189],{"categories":822},[192],{"categories":824},[],{"categories":826},[244],{"categories":828},[244],{"categories":830},[189],{"categories":832},[],{"categories":834},[192],{"categories":836},[192],{"categories":838},[189],{"categories":840},[213],{"categories":842},[251],{"categories":844},[195],{"categories":846},[],{"categories":848},[234],{"categories":850},[],{"categories":852},[192],{"categories":854},[],{"categories":856},[189],{"categories":858},[195],{"categories":860},[],{"categories":862},[506],{"categories":864},[237],{"categories":866},[244],{"categories":868},[251],{"categories":870},[244],{"categories":872},[195],{"categories":874},[],{"categories":876},[],{"categories":878},[195],{"categories":880},[186],{"categories":882},[195],{"categories":884},[198],{"categories":886},[189],{"categories":888},[],{"categories":890},[192],{"categories":892},[198],{"categories":894},[192],{"categories":896},[192],{"categories":898},[251],{"categories":900},[234],{"categories":902},[195],{"categories":904},[],{"categories":906},[],{"categories":908},[506],{"categories":910},[244],{"categories":912},[],{"categories":914},[195],{"categories":916},[192],{"categories":918},[234,192],{"categories":920},[186],{"categories":922},[],{"categories":924},[192],{"categories":926},[186],{"categories":928},[234],{"categories":930},[195],{"categories":932},[244],{"categories":934},[],{"categories":936},[192],{"categories":938},[],{"categories":940},[186],{"categories":942},[],{"categories":944},[195],{"categories":946},[198],{"categories":948},[192],{"categories":950},[192],{"categories":952},[234],{"categories":954},[195],{"categories":956},[506],{"categories":958},[234],{"categories":960},[195],{"categories":962},[192],{"categories":964},[192],{"categories":966},[192],{"categories":968},[213],{"categories":970},[],{"categories":972},[198],{"categories":974},[195],{"categories":976},[234],{"categories":978},[195],{"categories":980},[244],{"categories":982},[234],{"categories":984},[195],{"categories":986},[213],{"categories":988},[],{"categories":990},[192],{"categories":992},[234],{"categories":994},[192],{"categories":996},[186],{"categories":998},[213],{"categories":1000},[192],{"categories":1002},[251],{"categories":1004},[192],{"categories":1006},[192],{"categories":1008},[195],{"categories":1010},[195],{"categories":1012},[192],{"categories":1014},[195],{"categories":1016},[234],{"categories":1018},[192],{"categories":1020},[],{"categories":1022},[],{"categories":1024},[244],{"categories":1026},[],{"categories":1028},[186],{"categories":1030},[506],{"categories":1032},[],{"categories":1034},[186],{"categories":1036},[189],{"categories":1038},[251],{"categories":1040},[],{"categories":1042},[189],{"categories":1044},[],{"categories":1046},[],{"categories":1048},[],{"categories":1050},[],{"categories":1052},[],{"categories":1054},[192],{"categories":1056},[195],{"categories":1058},[506],{"categories":1060},[186],{"categories":1062},[192],{"categories":1064},[244],{"categories":1066},[198],{"categories":1068},[192],{"categories":1070},[251],{"categories":1072},[192],{"categories":1074},[192],{"categories":1076},[192],{"categories":1078},[192,186],{"categories":1080},[244],{"categories":1082},[244],{"categories":1084},[234],{"categories":1086},[192],{"categories":1088},[],{"categories":1090},[],{"categories":1092},[],{"categories":1094},[244],{"categories":1096},[237],{"categories":1098},[213],{"categories":1100},[234],{"categories":1102},[],{"categories":1104},[192],{"categories":1106},[192],{"categories":1108},[],{"categories":1110},[],{"categories":1112},[195],{"categories":1114},[192],{"categories":1116},[189],{"categories":1118},[],{"categories":1120},[186],{"categories":1122},[192],{"categories":1124},[186],{"categories":1126},[192],{"categories":1128},[244],{"categories":1130},[251],{"categories":1132},[192,234],{"categories":1134},[213],{"categories":1136},[234],{"categories":1138},[],{"categories":1140},[506],{"categories":1142},[234],{"categories":1144},[195],{"categories":1146},[],{"categories":1148},[],{"categories":1150},[],{"categories":1152},[],{"categories":1154},[244],{"categories":1156},[195],{"categories":1158},[195],{"categories":1160},[192],{"categories":1162},[192],{"categories":1164},[],{"categories":1166},[234],{"categories":1168},[],{"categories":1170},[],{"categories":1172},[195],{"categories":1174},[],{"categories":1176},[],{"categories":1178},[251],{"categories":1180},[251],{"categories":1182},[195],{"categories":1184},[],{"categories":1186},[192],{"categories":1188},[192],{"categories":1190},[244],{"categories":1192},[234],{"categories":1194},[234],{"categories":1196},[195],{"categories":1198},[186],{"categories":1200},[192],{"categories":1202},[234],{"categories":1204},[234],{"categories":1206},[195],{"categories":1208},[195],{"categories":1210},[192],{"categories":1212},[],{"categories":1214},[],{"categories":1216},[192],{"categories":1218},[195],{"categories":1220},[213],{"categories":1222},[244],{"categories":1224},[186],{"categories":1226},[192],{"categories":1228},[],{"categories":1230},[195],{"categories":1232},[195],{"categories":1234},[],{"categories":1236},[186],{"categories":1238},[192],{"categories":1240},[186],{"categories":1242},[186],{"categories":1244},[],{"categories":1246},[],{"categories":1248},[195],{"categories":1250},[195],{"categories":1252},[192],{"categories":1254},[192],{"categories":1256},[213],{"categories":1258},[237],{"categories":1260},[198],{"categories":1262},[213],{"categories":1264},[234],{"categories":1266},[],{"categories":1268},[213],{"categories":1270},[],{"categories":1272},[],{"categories":1274},[],{"categories":1276},[],{"categories":1278},[244],{"categories":1280},[237],{"categories":1282},[],{"categories":1284},[192],{"categories":1286},[192],{"categories":1288},[237],{"categories":1290},[244],{"categories":1292},[],{"categories":1294},[],{"categories":1296},[195],{"categories":1298},[213],{"categories":1300},[213],{"categories":1302},[195],{"categories":1304},[186],{"categories":1306},[192,506],{"categories":1308},[],{"categories":1310},[234],{"categories":1312},[186],{"categories":1314},[195],{"categories":1316},[234],{"categories":1318},[],{"categories":1320},[195],{"categories":1322},[195],{"categories":1324},[192],{"categories":1326},[251],{"categories":1328},[244],{"categories":1330},[234],{"categories":1332},[],{"categories":1334},[195],{"categories":1336},[192],{"categories":1338},[195],{"categories":1340},[195],{"categories":1342},[195],{"categories":1344},[251],{"categories":1346},[195],{"categories":1348},[192],{"categories":1350},[],{"categories":1352},[251],{"categories":1354},[213],{"categories":1356},[195],{"categories":1358},[],{"categories":1360},[],{"categories":1362},[192],{"categories":1364},[195],{"categories":1366},[213],{"categories":1368},[195],{"categories":1370},[],{"categories":1372},[],{"categories":1374},[],{"categories":1376},[195],{"categories":1378},[],{"categories":1380},[],{"categories":1382},[237],{"categories":1384},[192],{"categories":1386},[237],{"categories":1388},[213],{"categories":1390},[192],{"categories":1392},[192],{"categories":1394},[195],{"categories":1396},[192],{"categories":1398},[],{"categories":1400},[],{"categories":1402},[506],{"categories":1404},[],{"categories":1406},[],{"categories":1408},[186],{"categories":1410},[],{"categories":1412},[],{"categories":1414},[],{"categories":1416},[],{"categories":1418},[244],{"categories":1420},[213],{"categories":1422},[251],{"categories":1424},[189],{"categories":1426},[192],{"categories":1428},[192],{"categories":1430},[189],{"categories":1432},[],{"categories":1434},[234],{"categories":1436},[195],{"categories":1438},[189],{"categories":1440},[192],{"categories":1442},[192],{"categories":1444},[186],{"categories":1446},[],{"categories":1448},[186],{"categories":1450},[192],{"categories":1452},[251],{"categories":1454},[195],{"categories":1456},[213],{"categories":1458},[189],{"categories":1460},[192],{"categories":1462},[195],{"categories":1464},[],{"categories":1466},[192],{"categories":1468},[186],{"categories":1470},[192],{"categories":1472},[],{"categories":1474},[213],{"categories":1476},[192],{"categories":1478},[],{"categories":1480},[189],{"categories":1482},[192],{"categories":1484},[],{"categories":1486},[],{"categories":1488},[],{"categories":1490},[192],{"categories":1492},[],{"categories":1494},[506],{"categories":1496},[192],{"categories":1498},[],{"categories":1500},[192],{"categories":1502},[192],{"categories":1504},[192],{"categories":1506},[192,506],{"categories":1508},[192],{"categories":1510},[192],{"categories":1512},[234],{"categories":1514},[195],{"categories":1516},[],{"categories":1518},[195],{"categories":1520},[192],{"categories":1522},[192],{"categories":1524},[192],{"categories":1526},[186],{"categories":1528},[186],{"categories":1530},[244],{"categories":1532},[234],{"categories":1534},[195],{"categories":1536},[],{"categories":1538},[192],{"categories":1540},[213],{"categories":1542},[192],{"categories":1544},[189],{"categories":1546},[],{"categories":1548},[506],{"categories":1550},[234],{"categories":1552},[234],{"categories":1554},[195],{"categories":1556},[213],{"categories":1558},[195],{"categories":1560},[192],{"categories":1562},[],{"categories":1564},[192],{"categories":1566},[],{"categories":1568},[],{"categories":1570},[192],{"categories":1572},[192],{"categories":1574},[192],{"categories":1576},[195],{"categories":1578},[192],{"categories":1580},[],{"categories":1582},[237],{"categories":1584},[195],{"categories":1586},[],{"categories":1588},[192],{"categories":1590},[213],{"categories":1592},[],{"categories":1594},[234],{"categories":1596},[506],{"categories":1598},[213],{"categories":1600},[244],{"categories":1602},[244],{"categories":1604},[213],{"categories":1606},[213],{"categories":1608},[506],{"categories":1610},[],{"categories":1612},[213],{"categories":1614},[192],{"categories":1616},[186],{"categories":1618},[213],{"categories":1620},[],{"categories":1622},[237],{"categories":1624},[213],{"categories":1626},[244],{"categories":1628},[213],{"categories":1630},[506],{"categories":1632},[192],{"categories":1634},[192],{"categories":1636},[],{"categories":1638},[189],{"categories":1640},[],{"categories":1642},[],{"categories":1644},[192],{"categories":1646},[192],{"categories":1648},[192],{"categories":1650},[192],{"categories":1652},[],{"categories":1654},[237],{"categories":1656},[186],{"categories":1658},[],{"categories":1660},[192],{"categories":1662},[192],{"categories":1664},[506],{"categories":1666},[506],{"categories":1668},[],{"categories":1670},[195],{"categories":1672},[213],{"categories":1674},[213],{"categories":1676},[192],{"categories":1678},[195],{"categories":1680},[],{"categories":1682},[234],{"categories":1684},[192],{"categories":1686},[192],{"categories":1688},[],{"categories":1690},[],{"categories":1692},[506],{"categories":1694},[192],{"categories":1696},[244],{"categories":1698},[189],{"categories":1700},[192],{"categories":1702},[],{"categories":1704},[195],{"categories":1706},[186],{"categories":1708},[186],{"categories":1710},[],{"categories":1712},[192],{"categories":1714},[234],{"categories":1716},[195],{"categories":1718},[],{"categories":1720},[192],{"categories":1722},[192],{"categories":1724},[195],{"categories":1726},[],{"categories":1728},[195],{"categories":1730},[244],{"categories":1732},[],{"categories":1734},[192],{"categories":1736},[],{"categories":1738},[192],{"categories":1740},[],{"categories":1742},[192],{"categories":1744},[192],{"categories":1746},[],{"categories":1748},[192],{"categories":1750},[213],{"categories":1752},[192],{"categories":1754},[192],{"categories":1756},[186],{"categories":1758},[192],{"categories":1760},[213],{"categories":1762},[195],{"categories":1764},[],{"categories":1766},[192],{"categories":1768},[251],{"categories":1770},[],{"categories":1772},[],{"categories":1774},[],{"categories":1776},[186],{"categories":1778},[213],{"categories":1780},[195],{"categories":1782},[192],{"categories":1784},[234],{"categories":1786},[195],{"categories":1788},[],{"categories":1790},[195],{"categories":1792},[],{"categories":1794},[192],{"categories":1796},[195],{"categories":1798},[192],{"categories":1800},[],{"categories":1802},[192],{"categories":1804},[192],{"categories":1806},[213],{"categories":1808},[234],{"categories":1810},[195],{"categories":1812},[234],{"categories":1814},[189],{"categories":1816},[],{"categories":1818},[],{"categories":1820},[192],{"categories":1822},[186],{"categories":1824},[213],{"categories":1826},[],{"categories":1828},[],{"categories":1830},[244],{"categories":1832},[234],{"categories":1834},[],{"categories":1836},[192],{"categories":1838},[],{"categories":1840},[251],{"categories":1842},[192],{"categories":1844},[506],{"categories":1846},[244],{"categories":1848},[],{"categories":1850},[195],{"categories":1852},[192],{"categories":1854},[195],{"categories":1856},[195],{"categories":1858},[192],{"categories":1860},[],{"categories":1862},[186],{"categories":1864},[192],{"categories":1866},[189],{"categories":1868},[244],{"categories":1870},[234],{"categories":1872},[],{"categories":1874},[],{"categories":1876},[],{"categories":1878},[195],{"categories":1880},[234],{"categories":1882},[213],{"categories":1884},[192],{"categories":1886},[213],{"categories":1888},[234],{"categories":1890},[],{"categories":1892},[234],{"categories":1894},[213],{"categories":1896},[189],{"categories":1898},[192],{"categories":1900},[213],{"categories":1902},[251],{"categories":1904},[],{"categories":1906},[],{"categories":1908},[237],{"categories":1910},[192,244],{"categories":1912},[213],{"categories":1914},[192],{"categories":1916},[195],{"categories":1918},[195],{"categories":1920},[192],{"categories":1922},[],{"categories":1924},[244],{"categories":1926},[192],{"categories":1928},[237],{"categories":1930},[195],{"categories":1932},[251],{"categories":1934},[506],{"categories":1936},[],{"categories":1938},[186],{"categories":1940},[195],{"categories":1942},[195],{"categories":1944},[244],{"categories":1946},[192],{"categories":1948},[192],{"categories":1950},[],{"categories":1952},[],{"categories":1954},[],{"categories":1956},[506],{"categories":1958},[213],{"categories":1960},[192],{"categories":1962},[192],{"categories":1964},[192],{"categories":1966},[],{"categories":1968},[237],{"categories":1970},[189],{"categories":1972},[],{"categories":1974},[195],{"categories":1976},[506],{"categories":1978},[],{"categories":1980},[234],{"categories":1982},[234],{"categories":1984},[],{"categories":1986},[244],{"categories":1988},[234],{"categories":1990},[192],{"categories":1992},[],{"categories":1994},[213],{"categories":1996},[192],{"categories":1998},[234],{"categories":2000},[195],{"categories":2002},[213],{"categories":2004},[],{"categories":2006},[195],{"categories":2008},[234],{"categories":2010},[192],{"categories":2012},[],{"categories":2014},[192],{"categories":2016},[192],{"categories":2018},[506],{"categories":2020},[213],{"categories":2022},[237],{"categories":2024},[237],{"categories":2026},[],{"categories":2028},[],{"categories":2030},[],{"categories":2032},[195],{"categories":2034},[244],{"categories":2036},[244],{"categories":2038},[],{"categories":2040},[],{"categories":2042},[192],{"categories":2044},[],{"categories":2046},[195],{"categories":2048},[192],{"categories":2050},[],{"categories":2052},[192],{"categories":2054},[189],{"categories":2056},[192],{"categories":2058},[251],{"categories":2060},[195],{"categories":2062},[192],{"categories":2064},[244],{"categories":2066},[213],{"categories":2068},[195],{"categories":2070},[],{"categories":2072},[213],{"categories":2074},[195],{"categories":2076},[195],{"categories":2078},[],{"categories":2080},[189],{"categories":2082},[195],{"categories":2084},[],{"categories":2086},[192],{"categories":2088},[186],{"categories":2090},[213],{"categories":2092},[506],{"categories":2094},[195],{"categories":2096},[195],{"categories":2098},[186],{"categories":2100},[192],{"categories":2102},[],{"categories":2104},[],{"categories":2106},[234],{"categories":2108},[192,189],{"categories":2110},[],{"categories":2112},[186],{"categories":2114},[237],{"categories":2116},[192],{"categories":2118},[244],{"categories":2120},[192],{"categories":2122},[195],{"categories":2124},[192],{"categories":2126},[192],{"categories":2128},[213],{"categories":2130},[195],{"categories":2132},[],{"categories":2134},[],{"categories":2136},[195],{"categories":2138},[192],{"categories":2140},[506],{"categories":2142},[],{"categories":2144},[192],{"categories":2146},[195],{"categories":2148},[],{"categories":2150},[192],{"categories":2152},[251],{"categories":2154},[237],{"categories":2156},[195],{"categories":2158},[192],{"categories":2160},[506],{"categories":2162},[],{"categories":2164},[192],{"categories":2166},[251],{"categories":2168},[234],{"categories":2170},[192],{"categories":2172},[],{"categories":2174},[251],{"categories":2176},[213],{"categories":2178},[192],{"categories":2180},[192],{"categories":2182},[186],{"categories":2184},[],{"categories":2186},[],{"categories":2188},[234],{"categories":2190},[192],{"categories":2192},[237],{"categories":2194},[251],{"categories":2196},[251],{"categories":2198},[213],{"categories":2200},[],{"categories":2202},[],{"categories":2204},[192],{"categories":2206},[],{"categories":2208},[192,244],{"categories":2210},[213],{"categories":2212},[195],{"categories":2214},[244],{"categories":2216},[192],{"categories":2218},[186],{"categories":2220},[],{"categories":2222},[],{"categories":2224},[186],{"categories":2226},[251],{"categories":2228},[192],{"categories":2230},[],{"categories":2232},[234,192],{"categories":2234},[506],{"categories":2236},[186],{"categories":2238},[],{"categories":2240},[189],{"categories":2242},[189],{"categories":2244},[192],{"categories":2246},[244],{"categories":2248},[195],{"categories":2250},[213],{"categories":2252},[251],{"categories":2254},[234],{"categories":2256},[192],{"categories":2258},[192],{"categories":2260},[192],{"categories":2262},[186],{"categories":2264},[192],{"categories":2266},[195],{"categories":2268},[213],{"categories":2270},[],{"categories":2272},[],{"categories":2274},[237],{"categories":2276},[244],{"categories":2278},[192],{"categories":2280},[234],{"categories":2282},[237],{"categories":2284},[192],{"categories":2286},[192],{"categories":2288},[195],{"categories":2290},[195],{"categories":2292},[192,189],{"categories":2294},[],{"categories":2296},[234],{"categories":2298},[],{"categories":2300},[192],{"categories":2302},[213],{"categories":2304},[186],{"categories":2306},[186],{"categories":2308},[195],{"categories":2310},[192],{"categories":2312},[189],{"categories":2314},[244],{"categories":2316},[251],{"categories":2318},[],{"categories":2320},[213],{"categories":2322},[192],{"categories":2324},[192],{"categories":2326},[213],{"categories":2328},[244],{"categories":2330},[192],{"categories":2332},[195],{"categories":2334},[213],{"categories":2336},[192],{"categories":2338},[234],{"categories":2340},[192],{"categories":2342},[192],{"categories":2344},[506],{"categories":2346},[198],{"categories":2348},[195],{"categories":2350},[192],{"categories":2352},[213],{"categories":2354},[195],{"categories":2356},[251],{"categories":2358},[192],{"categories":2360},[],{"categories":2362},[192],{"categories":2364},[],{"categories":2366},[],{"categories":2368},[],{"categories":2370},[189],{"categories":2372},[192],{"categories":2374},[195],{"categories":2376},[213],{"categories":2378},[213],{"categories":2380},[213],{"categories":2382},[213],{"categories":2384},[],{"categories":2386},[186],{"categories":2388},[195],{"categories":2390},[213],{"categories":2392},[186],{"categories":2394},[195],{"categories":2396},[192],{"categories":2398},[192,195],{"categories":2400},[195],{"categories":2402},[506],{"categories":2404},[213],{"categories":2406},[213],{"categories":2408},[195],{"categories":2410},[192],{"categories":2412},[],{"categories":2414},[213],{"categories":2416},[251],{"categories":2418},[186],{"categories":2420},[192],{"categories":2422},[192],{"categories":2424},[],{"categories":2426},[244],{"categories":2428},[],{"categories":2430},[186],{"categories":2432},[195],{"categories":2434},[213],{"categories":2436},[192],{"categories":2438},[213],{"categories":2440},[186],{"categories":2442},[213],{"categories":2444},[213],{"categories":2446},[],{"categories":2448},[189],{"categories":2450},[195],{"categories":2452},[213],{"categories":2454},[213],{"categories":2456},[213],{"categories":2458},[213],{"categories":2460},[213],{"categories":2462},[213],{"categories":2464},[213],{"categories":2466},[213],{"categories":2468},[213],{"categories":2470},[213],{"categories":2472},[237],{"categories":2474},[186],{"categories":2476},[192],{"categories":2478},[192],{"categories":2480},[],{"categories":2482},[192,186],{"categories":2484},[],{"categories":2486},[195],{"categories":2488},[213],{"categories":2490},[195],{"categories":2492},[192],{"categories":2494},[192],{"categories":2496},[192],{"categories":2498},[192],{"categories":2500},[192],{"categories":2502},[195],{"categories":2504},[189],{"categories":2506},[234],{"categories":2508},[213],{"categories":2510},[192],{"categories":2512},[],{"categories":2514},[],{"categories":2516},[195],{"categories":2518},[234],{"categories":2520},[192],{"categories":2522},[],{"categories":2524},[],{"categories":2526},[251],{"categories":2528},[192],{"categories":2530},[],{"categories":2532},[],{"categories":2534},[186],{"categories":2536},[189],{"categories":2538},[192],{"categories":2540},[189],{"categories":2542},[234],{"categories":2544},[],{"categories":2546},[213],{"categories":2548},[],{"categories":2550},[234],{"categories":2552},[192],{"categories":2554},[251],{"categories":2556},[],{"categories":2558},[251],{"categories":2560},[],{"categories":2562},[],{"categories":2564},[195],{"categories":2566},[],{"categories":2568},[189],{"categories":2570},[186],{"categories":2572},[234],{"categories":2574},[244],{"categories":2576},[],{"categories":2578},[],{"categories":2580},[192],{"categories":2582},[186],{"categories":2584},[251],{"categories":2586},[],{"categories":2588},[195],{"categories":2590},[195],{"categories":2592},[213],{"categories":2594},[192],{"categories":2596},[195],{"categories":2598},[192],{"categories":2600},[195],{"categories":2602},[192],{"categories":2604},[198],{"categories":2606},[213],{"categories":2608},[],{"categories":2610},[251],{"categories":2612},[244],{"categories":2614},[195],{"categories":2616},[],{"categories":2618},[192],{"categories":2620},[195],{"categories":2622},[189],{"categories":2624},[186],{"categories":2626},[192],{"categories":2628},[234],{"categories":2630},[244],{"categories":2632},[244],{"categories":2634},[192],{"categories":2636},[237],{"categories":2638},[192],{"categories":2640},[195],{"categories":2642},[189],{"categories":2644},[195],{"categories":2646},[192],{"categories":2648},[192],{"categories":2650},[195],{"categories":2652},[213],{"categories":2654},[],{"categories":2656},[186],{"categories":2658},[192],{"categories":2660},[195],{"categories":2662},[192],{"categories":2664},[192],{"categories":2666},[],{"categories":2668},[234],{"categories":2670},[189],{"categories":2672},[213],{"categories":2674},[192],{"categories":2676},[192],{"categories":2678},[234],{"categories":2680},[251],{"categories":2682},[237],{"categories":2684},[192],{"categories":2686},[213],{"categories":2688},[192],{"categories":2690},[195],{"categories":2692},[506],{"categories":2694},[192],{"categories":2696},[195],{"categories":2698},[237],{"categories":2700},[],{"categories":2702},[195],{"categories":2704},[244],{"categories":2706},[234],{"categories":2708},[192],{"categories":2710},[186],{"categories":2712},[189],{"categories":2714},[244],{"categories":2716},[],{"categories":2718},[195],{"categories":2720},[192],{"categories":2722},[],{"categories":2724},[213],{"categories":2726},[],{"categories":2728},[213],{"categories":2730},[192],{"categories":2732},[195],{"categories":2734},[195],{"categories":2736},[195],{"categories":2738},[],{"categories":2740},[],{"categories":2742},[192],{"categories":2744},[192],{"categories":2746},[],{"categories":2748},[234],{"categories":2750},[195],{"categories":2752},[251],{"categories":2754},[186],{"categories":2756},[],{"categories":2758},[],{"categories":2760},[213],{"categories":2762},[244],{"categories":2764},[192],{"categories":2766},[192],{"categories":2768},[192],{"categories":2770},[244],{"categories":2772},[213],{"categories":2774},[234],{"categories":2776},[192],{"categories":2778},[192],{"categories":2780},[192],{"categories":2782},[213],{"categories":2784},[192],{"categories":2786},[213],{"categories":2788},[195],{"categories":2790},[195],{"categories":2792},[244],{"categories":2794},[195],{"categories":2796},[192],{"categories":2798},[244],{"categories":2800},[234],{"categories":2802},[],{"categories":2804},[195],{"categories":2806},[],{"categories":2808},[],{"categories":2810},[189],{"categories":2812},[192],{"categories":2814},[195],{"categories":2816},[186],{"categories":2818},[195],{"categories":2820},[251],{"categories":2822},[],{"categories":2824},[195],{"categories":2826},[],{"categories":2828},[186],{"categories":2830},[195],{"categories":2832},[],{"categories":2834},[195],{"categories":2836},[192],{"categories":2838},[213],{"categories":2840},[192],{"categories":2842},[195],{"categories":2844},[213],{"categories":2846},[195],{"categories":2848},[244],{"categories":2850},[234],{"categories":2852},[186],{"categories":2854},[],{"categories":2856},[195],{"categories":2858},[234],{"categories":2860},[213],{"categories":2862},[192],{"categories":2864},[234],{"categories":2866},[186],{"categories":2868},[],{"categories":2870},[195],{"categories":2872},[195],{"categories":2874},[192],{"categories":2876},[],{"categories":2878},[195],{"categories":2880},[198],{"categories":2882},[213],{"categories":2884},[195],{"categories":2886},[189],{"categories":2888},[],{"categories":2890},[192],{"categories":2892},[198],{"categories":2894},[192],{"categories":2896},[195],{"categories":2898},[213],{"categories":2900},[186],{"categories":2902},[506],{"categories":2904},[192],{"categories":2906},[192],{"categories":2908},[192],{"categories":2910},[213],{"categories":2912},[189],{"categories":2914},[192],{"categories":2916},[234],{"categories":2918},[213],{"categories":2920},[506],{"categories":2922},[192],{"categories":2924},[],{"categories":2926},[],{"categories":2928},[506],{"categories":2930},[237],{"categories":2932},[195],{"categories":2934},[195],{"categories":2936},[213],{"categories":2938},[192],{"categories":2940},[186],{"categories":2942},[234],{"categories":2944},[195],{"categories":2946},[192],{"categories":2948},[251],{"categories":2950},[192],{"categories":2952},[195],{"categories":2954},[],{"categories":2956},[192],{"categories":2958},[192],{"categories":2960},[213],{"categories":2962},[186],{"categories":2964},[],{"categories":2966},[192],{"categories":2968},[192],{"categories":2970},[244],{"categories":2972},[234],{"categories":2974},[192,195],{"categories":2976},[251,189],{"categories":2978},[192],{"categories":2980},[],{"categories":2982},[195],{"categories":2984},[],{"categories":2986},[244],{"categories":2988},[192],{"categories":2990},[213],{"categories":2992},[],{"categories":2994},[195],{"categories":2996},[],{"categories":2998},[195],{"categories":3000},[186],{"categories":3002},[195],{"categories":3004},[192],{"categories":3006},[506],{"categories":3008},[251],{"categories":3010},[189],{"categories":3012},[189],{"categories":3014},[186],{"categories":3016},[186],{"categories":3018},[192],{"categories":3020},[195],{"categories":3022},[192],{"categories":3024},[192],{"categories":3026},[186],{"categories":3028},[192],{"categories":3030},[251],{"categories":3032},[213],{"categories":3034},[192],{"categories":3036},[195],{"categories":3038},[192],{"categories":3040},[],{"categories":3042},[244],{"categories":3044},[],{"categories":3046},[195],{"categories":3048},[186],{"categories":3050},[],{"categories":3052},[506],{"categories":3054},[192],{"categories":3056},[],{"categories":3058},[213],{"categories":3060},[195],{"categories":3062},[244],{"categories":3064},[192],{"categories":3066},[195],{"categories":3068},[244],{"categories":3070},[195],{"categories":3072},[213],{"categories":3074},[186],{"categories":3076},[213],{"categories":3078},[244],{"categories":3080},[192],{"categories":3082},[234],{"categories":3084},[192],{"categories":3086},[192],{"categories":3088},[192],{"categories":3090},[192],{"categories":3092},[195],{"categories":3094},[192],{"categories":3096},[195],{"categories":3098},[192],{"categories":3100},[186],{"categories":3102},[192],{"categories":3104},[195],{"categories":3106},[234],{"categories":3108},[186],{"categories":3110},[195],{"categories":3112},[234],{"categories":3114},[],{"categories":3116},[192],{"categories":3118},[192],{"categories":3120},[244],{"categories":3122},[],{"categories":3124},[195],{"categories":3126},[251],{"categories":3128},[192],{"categories":3130},[213],{"categories":3132},[251],{"categories":3134},[195],{"categories":3136},[189],{"categories":3138},[189],{"categories":3140},[192],{"categories":3142},[186],{"categories":3144},[],{"categories":3146},[192],{"categories":3148},[],{"categories":3150},[186],{"categories":3152},[192],{"categories":3154},[195],{"categories":3156},[195],{"categories":3158},[],{"categories":3160},[244],{"categories":3162},[244],{"categories":3164},[251],{"categories":3166},[234],{"categories":3168},[],{"categories":3170},[192],{"categories":3172},[186],{"categories":3174},[192],{"categories":3176},[244],{"categories":3178},[186],{"categories":3180},[213],{"categories":3182},[213],{"categories":3184},[],{"categories":3186},[213],{"categories":3188},[195],{"categories":3190},[234],{"categories":3192},[237],{"categories":3194},[192],{"categories":3196},[],{"categories":3198},[213],{"categories":3200},[244],{"categories":3202},[189],{"categories":3204},[192],{"categories":3206},[186],{"categories":3208},[506],{"categories":3210},[186],{"categories":3212},[],{"categories":3214},[],{"categories":3216},[213],{"categories":3218},[],{"categories":3220},[195],{"categories":3222},[195],{"categories":3224},[195],{"categories":3226},[],{"categories":3228},[192],{"categories":3230},[],{"categories":3232},[213],{"categories":3234},[186],{"categories":3236},[234],{"categories":3238},[192],{"categories":3240},[213],{"categories":3242},[213],{"categories":3244},[],{"categories":3246},[213],{"categories":3248},[186],{"categories":3250},[192],{"categories":3252},[],{"categories":3254},[195],{"categories":3256},[195],{"categories":3258},[186],{"categories":3260},[],{"categories":3262},[],{"categories":3264},[],{"categories":3266},[234],{"categories":3268},[195],{"categories":3270},[192],{"categories":3272},[],{"categories":3274},[],{"categories":3276},[],{"categories":3278},[234],{"categories":3280},[],{"categories":3282},[186],{"categories":3284},[],{"categories":3286},[],{"categories":3288},[234],{"categories":3290},[192],{"categories":3292},[213],{"categories":3294},[],{"categories":3296},[251],{"categories":3298},[213],{"categories":3300},[251],{"categories":3302},[192],{"categories":3304},[],{"categories":3306},[],{"categories":3308},[195],{"categories":3310},[],{"categories":3312},[],{"categories":3314},[195],{"categories":3316},[192],{"categories":3318},[],{"categories":3320},[195],{"categories":3322},[213],{"categories":3324},[251],{"categories":3326},[237],{"categories":3328},[195],{"categories":3330},[195],{"categories":3332},[],{"categories":3334},[],{"categories":3336},[],{"categories":3338},[213],{"categories":3340},[],{"categories":3342},[],{"categories":3344},[234],{"categories":3346},[186],{"categories":3348},[],{"categories":3350},[189],{"categories":3352},[251],{"categories":3354},[192],{"categories":3356},[244],{"categories":3358},[186],{"categories":3360},[237],{"categories":3362},[189],{"categories":3364},[244],{"categories":3366},[],{"categories":3368},[],{"categories":3370},[195],{"categories":3372},[186],{"categories":3374},[234],{"categories":3376},[186],{"categories":3378},[195],{"categories":3380},[506],{"categories":3382},[195],{"categories":3384},[],{"categories":3386},[192],{"categories":3388},[213],{"categories":3390},[244],{"categories":3392},[],{"categories":3394},[234],{"categories":3396},[213],{"categories":3398},[186],{"categories":3400},[195],{"categories":3402},[192],{"categories":3404},[189],{"categories":3406},[195,506],{"categories":3408},[195],{"categories":3410},[244],{"categories":3412},[192],{"categories":3414},[237],{"categories":3416},[251],{"categories":3418},[195],{"categories":3420},[],{"categories":3422},[195],{"categories":3424},[192],{"categories":3426},[189],{"categories":3428},[],{"categories":3430},[],{"categories":3432},[192],{"categories":3434},[237],{"categories":3436},[192],{"categories":3438},[],{"categories":3440},[213],{"categories":3442},[],{"categories":3444},[213],{"categories":3446},[244],{"categories":3448},[195],{"categories":3450},[192],{"categories":3452},[251],{"categories":3454},[244],{"categories":3456},[],{"categories":3458},[213],{"categories":3460},[192],{"categories":3462},[],{"categories":3464},[192],{"categories":3466},[195],{"categories":3468},[192],{"categories":3470},[195],{"categories":3472},[192],{"categories":3474},[192],{"categories":3476},[192],{"categories":3478},[192],{"categories":3480},[189],{"categories":3482},[],{"categories":3484},[198],{"categories":3486},[213],{"categories":3488},[192],{"categories":3490},[],{"categories":3492},[244],{"categories":3494},[192],{"categories":3496},[192],{"categories":3498},[195],{"categories":3500},[213],{"categories":3502},[192],{"categories":3504},[192],{"categories":3506},[189],{"categories":3508},[195],{"categories":3510},[234],{"categories":3512},[],{"categories":3514},[237],{"categories":3516},[192],{"categories":3518},[],{"categories":3520},[213],{"categories":3522},[251],{"categories":3524},[],{"categories":3526},[],{"categories":3528},[213],{"categories":3530},[213],{"categories":3532},[251],{"categories":3534},[186],{"categories":3536},[195],{"categories":3538},[195],{"categories":3540},[192],{"categories":3542},[189],{"categories":3544},[],{"categories":3546},[],{"categories":3548},[213],{"categories":3550},[237],{"categories":3552},[244],{"categories":3554},[195],{"categories":3556},[234],{"categories":3558},[237],{"categories":3560},[237],{"categories":3562},[],{"categories":3564},[213],{"categories":3566},[192],{"categories":3568},[192],{"categories":3570},[244],{"categories":3572},[],{"categories":3574},[213],{"categories":3576},[213],{"categories":3578},[213],{"categories":3580},[],{"categories":3582},[195],{"categories":3584},[192],{"categories":3586},[],{"categories":3588},[186],{"categories":3590},[189],{"categories":3592},[],{"categories":3594},[192],{"categories":3596},[192],{"categories":3598},[],{"categories":3600},[244],{"categories":3602},[],{"categories":3604},[],{"categories":3606},[],{"categories":3608},[],{"categories":3610},[192],{"categories":3612},[213],{"categories":3614},[],{"categories":3616},[],{"categories":3618},[192],{"categories":3620},[192],{"categories":3622},[192],{"categories":3624},[237],{"categories":3626},[192],{"categories":3628},[237],{"categories":3630},[],{"categories":3632},[237],{"categories":3634},[237],{"categories":3636},[506],{"categories":3638},[195],{"categories":3640},[244],{"categories":3642},[],{"categories":3644},[],{"categories":3646},[237],{"categories":3648},[244],{"categories":3650},[244],{"categories":3652},[244],{"categories":3654},[],{"categories":3656},[186],{"categories":3658},[244],{"categories":3660},[244],{"categories":3662},[186],{"categories":3664},[244],{"categories":3666},[189],{"categories":3668},[244],{"categories":3670},[244],{"categories":3672},[244],{"categories":3674},[237],{"categories":3676},[213],{"categories":3678},[213],{"categories":3680},[192],{"categories":3682},[244],{"categories":3684},[237],{"categories":3686},[506],{"categories":3688},[237],{"categories":3690},[237],{"categories":3692},[237],{"categories":3694},[],{"categories":3696},[189],{"categories":3698},[],{"categories":3700},[506],{"categories":3702},[244],{"categories":3704},[244],{"categories":3706},[244],{"categories":3708},[195],{"categories":3710},[213,189],{"categories":3712},[237],{"categories":3714},[],{"categories":3716},[],{"categories":3718},[237],{"categories":3720},[],{"categories":3722},[237],{"categories":3724},[213],{"categories":3726},[195],{"categories":3728},[],{"categories":3730},[244],{"categories":3732},[192],{"categories":3734},[234],{"categories":3736},[],{"categories":3738},[192],{"categories":3740},[],{"categories":3742},[213],{"categories":3744},[186],{"categories":3746},[237],{"categories":3748},[],{"categories":3750},[244],{"categories":3752},[213],[3754,3815,3954,4032],{"id":3755,"title":3756,"ai":3757,"body":3762,"categories":3790,"created_at":132,"date_modified":132,"description":125,"extension":133,"faq":132,"featured":134,"kicker_label":132,"meta":3791,"navigation":166,"path":3802,"published_at":3803,"question":132,"scraped_at":3804,"seo":3805,"sitemap":3806,"source_id":3807,"source_name":3808,"source_type":173,"source_url":3809,"stem":3810,"tags":3811,"thumbnail_url":132,"tldr":3812,"tweet":132,"unknown_tags":3813,"__hash__":3814},"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":3758,"output_tokens":3759,"processing_time_ms":3760,"cost_usd":3761},6300,1652,21300,0.00157915,{"type":14,"value":3763,"toc":3785},[3764,3768,3771,3775,3778,3782],[17,3765,3767],{"id":3766},"tackling-agentic-inference-bottlenecks","Tackling Agentic Inference Bottlenecks",[22,3769,3770],{},"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.",[17,3772,3774],{"id":3773},"architectural-edges-for-speed-and-safety","Architectural Edges for Speed and Safety",[22,3776,3777],{},"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.",[17,3779,3781],{"id":3780},"benchmark-dominance-on-real-workloads","Benchmark Dominance on Real Workloads",[22,3783,3784],{},"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":125,"searchDepth":126,"depth":126,"links":3786},[3787,3788,3789],{"id":3766,"depth":126,"text":3767},{"id":3773,"depth":126,"text":3774},{"id":3780,"depth":126,"text":3781},[192],{"content_references":3792,"triage":3799},[3793,3796],{"type":153,"title":3794,"url":3795,"context":145},"TokenSpeed","https:\u002F\u002Fgithub.com\u002Flightseekorg\u002Ftokenspeed",{"type":138,"title":3797,"url":3798,"context":156},"LightSeek TokenSpeed Technical Details","https:\u002F\u002Flightseek.org\u002Fblog\u002Flightseek-tokenspeed.html",{"relevance":162,"novelty":163,"quality":162,"actionability":163,"composite":3800,"reasoning":3801},3.6,"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":3756,"description":125},{"loc":3802},"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",[177,178,179],"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":3816,"title":3817,"ai":3818,"body":3823,"categories":3927,"created_at":132,"date_modified":132,"description":125,"extension":133,"faq":132,"featured":134,"kicker_label":132,"meta":3928,"navigation":166,"path":3942,"published_at":3943,"question":132,"scraped_at":3944,"seo":3945,"sitemap":3946,"source_id":3947,"source_name":3937,"source_type":173,"source_url":3948,"stem":3949,"tags":3950,"thumbnail_url":132,"tldr":3951,"tweet":132,"unknown_tags":3952,"__hash__":3953},"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":3819,"output_tokens":3820,"processing_time_ms":3821,"cost_usd":3822},8403,2068,28390,0.00242375,{"type":14,"value":3824,"toc":3920},[3825,3829,3832,3835,3839,3842,3845,3848,3852,3855,3858,3861,3865,3868,3871,3874,3893,3897],[17,3826,3828],{"id":3827},"open-source-as-ais-innovation-backbone","Open Source as AI's Innovation Backbone",[22,3830,3831],{},"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.",[22,3833,3834],{},"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.",[17,3836,3838],{"id":3837},"secure-vs-securable-core-security-distinction","Secure vs. Securable: Core Security Distinction",[22,3840,3841],{},"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.",[22,3843,3844],{},"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.",[22,3846,3847],{},"Panel agrees on trust: opacity breeds blind faith, not security. Open source invites scrutiny, but demands proactive policy.",[17,3849,3851],{"id":3850},"model-access-bad-actors-and-emerging-threats","Model Access, Bad Actors, and Emerging Threats",[22,3853,3854],{},"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.",[22,3856,3857],{},"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.",[22,3859,3860],{},"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.",[17,3862,3864],{"id":3863},"using-ai-to-secure-ai-and-forward-outlook","Using AI to Secure AI and Forward Outlook",[22,3866,3867],{},"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.",[22,3869,3870],{},"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.",[22,3872,3873],{},"Notable quotes:",[3875,3876,3877,3881,3884,3887,3890],"ul",{},[3878,3879,3880],"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.\"",[3878,3882,3883],{},"Jeff Crume: \"Linux is a good example of a system that is securable, but in and of itself is not necessarily secure.\"",[3878,3885,3886],{},"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.\"",[3878,3888,3889],{},"Jeff Crume: \"Security through obscurity is not an effective model... the only thing about a crypto system that should be secret are the keys.\"",[3878,3891,3892],{},"Martin Keen: \"Open source is foundational to everything in AI now even if we're talking about models that were actually frontier closed models.\"",[17,3894,3896],{"id":3895},"key-takeaways","Key Takeaways",[3875,3898,3899,3902,3905,3908,3911,3914,3917],{},[3878,3900,3901],{},"Prioritize 'securable' open source projects with strong security contribution policies and update cadences—avoid vibe-check fails.",[3878,3903,3904],{},"Distinguish open code (innovation accelerator) from open weights (misuse risk)—use guardrails for models, sandboxes for agents.",[3878,3906,3907],{},"Leverage AI for vuln scanning on open or closed code; reverse-engineering erodes proprietary edges.",[3878,3909,3910],{},"Blend approaches: Open standards (MCP, skill.md) enhance closed models; run open weights on your infra for flexibility.",[3878,3912,3913],{},"Build trust via transparency and controls, not secrecy—bad actors leak anyway; focus on implementation.",[3878,3915,3916],{},"For agents, treat all inputs as untrusted code—rethink textual data assumptions.",[3878,3918,3919],{},"Expect open models to trail but catch frontier capabilities quickly via collaborative innovation.",{"title":125,"searchDepth":126,"depth":126,"links":3921},[3922,3923,3924,3925,3926],{"id":3827,"depth":126,"text":3828},{"id":3837,"depth":126,"text":3838},{"id":3850,"depth":126,"text":3851},{"id":3863,"depth":126,"text":3864},{"id":3895,"depth":126,"text":3896},[192],{"content_references":3929,"triage":3939},[3930,3935],{"type":3931,"title":3932,"author":3933,"url":3934,"context":145},"podcast","Security Intelligence x Mixture of Experts Crossover Episode","IBM Technology (Matt Kazinski host)","https:\u002F\u002Fibm.biz\u002F~sTfk9xICA",{"type":3931,"title":3936,"author":3937,"url":3938,"context":156},"Mixture of Experts","IBM Technology","https:\u002F\u002Fibm.biz\u002F~SMOMF0sqx",{"relevance":163,"novelty":163,"quality":162,"actionability":126,"composite":3940,"reasoning":3941},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":3817,"description":125},{"loc":3942},"e15ac1bd93fe2101","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=PJGIWDW_W2A","summaries\u002F9cf4eabf30c8f73e-open-source-ai-innovation-engine-or-security-risk-summary",[179,177,178],"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":3955,"title":3956,"ai":3957,"body":3962,"categories":4009,"created_at":132,"date_modified":132,"description":125,"extension":133,"faq":132,"featured":134,"kicker_label":132,"meta":4010,"navigation":166,"path":4019,"published_at":4020,"question":132,"scraped_at":4021,"seo":4022,"sitemap":4023,"source_id":4024,"source_name":4025,"source_type":173,"source_url":4026,"stem":4027,"tags":4028,"thumbnail_url":132,"tldr":4029,"tweet":132,"unknown_tags":4030,"__hash__":4031},"summaries\u002Fsummaries\u002Ffbbbc098d7e53ea7-hermes-agent-persists-learning-across-sessions-summary.md","Hermes Agent Persists Learning Across Sessions",{"provider":7,"model":8,"input_tokens":3958,"output_tokens":3959,"processing_time_ms":3960,"cost_usd":3961},3887,1356,12936,0.0014329,{"type":14,"value":3963,"toc":4004},[3964,3968,3971,3975,3983,3994,3997,4001],[17,3965,3967],{"id":3966},"session-amnesia-limits-current-ai-agents","Session Amnesia Limits Current AI Agents",[22,3969,3970],{},"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.",[17,3972,3974],{"id":3973},"hermes-builds-reusable-knowledge-via-learning-loop","Hermes Builds Reusable Knowledge via Learning Loop",[22,3976,3977,3978,3982],{},"Hermes Agent, an open-source project by Nous Research, embeds persistence from the ground up. Its core mechanism is a ",[3979,3980,3981],"em",{},"learning loop"," that:",[3875,3984,3985,3988,3991],{},[3878,3986,3987],{},"Records what worked in interactions.",[3878,3989,3990],{},"Distills those into reusable procedures.",[3878,3992,3993],{},"Automatically loads relevant procedures for matching future problems.",[22,3995,3996],{},"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.",[17,3998,4000],{"id":3999},"practical-value-for-ai-agent-builders","Practical Value for AI Agent Builders",[22,4002,4003],{},"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":125,"searchDepth":126,"depth":126,"links":4005},[4006,4007,4008],{"id":3966,"depth":126,"text":3967},{"id":3973,"depth":126,"text":3974},{"id":3999,"depth":126,"text":4000},[192],{"content_references":4011,"triage":4016},[4012],{"type":153,"title":4013,"author":4014,"url":4015,"context":145},"Hermes Agent","Nous Research","https:\u002F\u002Fnousresearch.com\u002F",{"relevance":161,"novelty":162,"quality":162,"actionability":162,"composite":4017,"reasoning":4018},4.35,"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":3956,"description":125},{"loc":4019},"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",[178,177,179],"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",{"id":4033,"title":4034,"ai":4035,"body":4040,"categories":4071,"created_at":132,"date_modified":132,"description":125,"extension":133,"faq":132,"featured":134,"kicker_label":132,"meta":4072,"navigation":166,"path":4085,"published_at":4086,"question":132,"scraped_at":4087,"seo":4088,"sitemap":4089,"source_id":4090,"source_name":3808,"source_type":173,"source_url":4091,"stem":4092,"tags":4093,"thumbnail_url":132,"tldr":4094,"tweet":132,"unknown_tags":4095,"__hash__":4096},"summaries\u002Fsummaries\u002Faa13e74f3ceda7e1-kimi-k2-6-open-moe-model-tops-agentic-coding-bench-summary.md","Kimi K2.6: Open MoE Model Tops Agentic Coding Benchmarks",{"provider":7,"model":8,"input_tokens":4036,"output_tokens":4037,"processing_time_ms":4038,"cost_usd":4039},6613,2117,17139,0.0023586,{"type":14,"value":4041,"toc":4066},[4042,4046,4049,4052,4056,4059,4063],[17,4043,4045],{"id":4044},"efficiently-deploy-1t-param-multimodal-moe-for-agentic-workloads","Efficiently Deploy 1T-Param Multimodal MoE for Agentic Workloads",[22,4047,4048],{},"Kimi K2.6 activates only 32B of its 1T parameters per token using a Mixture-of-Experts architecture with 384 experts (8 routed per token plus 1 shared), 61 layers, and Multi-head Latent Attention. Native vision via 400M-param MoonViT handles images\u002Fvideos alongside 256K-token context and 160K vocab. Run on vLLM, SGLang, or KTransformers with transformers >=4.57.1 \u003C5.0.0; reuse K2.5 configs. API offers Thinking mode (temp=1.0, chain-of-thought for coding\u002Fagents, enable preserve_thinking for multi-turn state) or Instant mode (disable via {'thinking': {'type': 'disabled'}} or chat_template_kwargs {\"thinking\": False}, temp=0.6 top-p=0.95). Weights on Hugging Face under Modified MIT.",[22,4050,4051],{},"Leads HLE-Full with tools at 54.0 (vs GPT-5.4 52.1, Claude Opus 4.6 53.0), SWE-Bench Pro 58.6 (vs GPT-5.4 57.7), Terminal-Bench 2.0 66.7, LiveCodeBench v6 89.6, BrowseComp swarm 86.3, DeepSearchQA 92.5 f1.",[17,4053,4055],{"id":4054},"execute-13-hour-autonomous-coding-with-4k-tool-calls","Execute 13-Hour Autonomous Coding with 4K+ Tool Calls",[22,4057,4058],{},"For long-horizon tasks, K2.6 resolves GitHub issues and optimizes code over hours without oversight. Example: Downloaded\u002Fdeployed Qwen3.5-0.8B on Mac, implemented Zig inference, iterated 14x over 12+ hours\u002F4K tool calls to boost throughput 15→193 tokens\u002Fsec (20% > LM Studio). Another: Refactored 8-year exchange-core engine (1K+ tool calls, 13 hours, 4K+ LOC changes); analyzed flame graphs, shifted topology 4ME+2RE→2ME+1RE for 185% medium throughput (0.43→1.24 MT\u002Fs) and 133% perf (1.23→2.86 MT\u002Fs). Internal Kimi Code Bench and Claw Bench show gains in coding, research, scheduling, memory over K2.5; RL team ran 5-day proactive agent for ops.",[17,4060,4062],{"id":4061},"scale-complex-tasks-via-300-sub-agent-swarms-and-claw-groups","Scale Complex Tasks via 300-Sub-Agent Swarms and Claw Groups",[22,4064,4065],{},"Agent Swarm decomposes tasks across 300 heterogeneous sub-agents\u002F4K steps (up from K2.5's 100\u002F1.5K), parallelizing search\u002Fresearch\u002Fwriting\u002Fgeneration for docs\u002Fsites\u002Fslides\u002Fspreadsheets. Skills feature ingests PDF\u002Fspreadsheet\u002Fslide\u002FWord to extract structure\u002Fstyle as reusable templates—e.g., CV→100 customized resumes; 30 LA stores→landing pages; astrophysics paper→40-page report +20K-entry dataset\u002F14 charts. Claw Groups (preview) coordinates external agents\u002Fhumans: K2.6 matches tasks by skills\u002Ftools, reassigns on failure, manages lifecycle. Used internally for content\u002FDemo\u002Fbenchmark\u002Fsocial\u002Fvideo production. Build persistent agents like OpenClaw\u002FHermes for 24\u002F7 cross-app ops.",{"title":125,"searchDepth":126,"depth":126,"links":4067},[4068,4069,4070],{"id":4044,"depth":126,"text":4045},{"id":4054,"depth":126,"text":4055},{"id":4061,"depth":126,"text":4062},[213],{"content_references":4073,"triage":4083},[4074,4077,4080],{"type":153,"title":4075,"url":4076,"context":145},"Kimi K2.6","https:\u002F\u002Fhuggingface.co\u002Fmoonshotai\u002FKimi-K2.6",{"type":153,"title":4078,"url":4079,"context":145},"Moonshot AI API","https:\u002F\u002Fplatform.moonshot.ai\u002F",{"type":138,"title":4081,"url":4082,"context":145},"Kimi K2.6 Technical Details","https:\u002F\u002Fwww.kimi.com\u002Fblog\u002Fkimi-k2-6",{"relevance":163,"novelty":163,"quality":162,"actionability":126,"composite":3940,"reasoning":4084},"Category: AI & LLMs. The article discusses the Kimi K2.6 model's capabilities in autonomous coding and agentic workloads, which is relevant to AI product builders. However, it lacks practical guidance or frameworks that the audience could directly apply in their work.","\u002Fsummaries\u002Faa13e74f3ceda7e1-kimi-k2-6-open-moe-model-tops-agentic-coding-bench-summary","2026-04-21 01:58:50","2026-04-21 15:26:54",{"title":4034,"description":125},{"loc":4085},"aa13e74f3ceda7e1","https:\u002F\u002Fwww.marktechpost.com\u002F2026\u002F04\u002F20\u002Fmoonshot-ai-releases-kimi-k2-6-with-long-horizon-coding-agent-swarm-scaling-to-300-sub-agents-and-4000-coordinated-steps\u002F","summaries\u002Faa13e74f3ceda7e1-kimi-k2-6-open-moe-model-tops-agentic-coding-bench-summary",[177,178,179],"Moonshot's 1T-param MoE Kimi K2.6 open-sources native multimodal agents that excel at 13-hour autonomous coding (185% throughput gains) and scale to 300 sub-agents over 4,000 steps, deployable via vLLM.",[],"TE7-oWRW_mPQDwC6Uxlkd32oGDIcC86i12EI3hfNl1A"]