{"generated_at":"2026-09-18T17:05:01Z","license":"CC-BY-4.0","attribution":"Bertrande, https://bertrande.com","measured_by":"Hugging Face router, https://router.huggingface.co/v1/models","note":"Time to first token and output throughput, MEASURED BY THE HUGGING FACE ROUTER on the traffic it sends, not by us and not by the host. They are not what you would measure calling a host yourself: neither the network path nor the load is the same. We read once a day and keep every reading. The median here is computed from our own dated readings, which is the only derivation in this file; readings_count and the period say how much it rests on, and a median over one reading is one reading. A row exists only where we already track that host's price for that model, so that no page is created carrying speed and no cost.","fields":["seller","model_slug","model_name","measure","unit","median","readings_count","min_seen","max_seen","first_reading","last_reading"],"count":360,"readings":[["baseten","deepseek-v4","DeepSeek V4 Pro (on Baseten)","latency_ttft_ms","ms",704.8,17,427.8,1438.4,"2026-09-02","2026-09-18"],["baseten","deepseek-v4","DeepSeek V4 Pro (on Baseten)","throughput_tps","tokens per second",87.317,17,57.841,106.039,"2026-09-02","2026-09-18"],["baseten","deepseek-v4-flash-0731","DeepSeek-V4-Flash-0731 (on Baseten)","latency_ttft_ms","ms",521.4,17,342.8,1919.0,"2026-09-02","2026-09-18"],["baseten","deepseek-v4-flash-0731","DeepSeek-V4-Flash-0731 (on Baseten)","throughput_tps","tokens per second",113.165,17,35.474,179.344,"2026-09-02","2026-09-18"],["baseten","deepseek-v4-pro-0813","DeepSeek V4 Pro 0813 (on Baseten)","latency_ttft_ms","ms",687.8,16,388.0,1278.4,"2026-09-03","2026-09-18"],["baseten","deepseek-v4-pro-0813","DeepSeek V4 Pro 0813 (on Baseten)","throughput_tps","tokens per second",81.294,16,47.272,105.831,"2026-09-03","2026-09-18"],["baseten","deepseek-v41-flash","DeepSeek V4.1 Flash (on Baseten)","latency_ttft_ms","ms",454.7,2,397.8,511.6,"2026-09-16","2026-09-17"],["baseten","deepseek-v41-flash","DeepSeek V4.1 Flash (on Baseten)","throughput_tps","tokens per second",161.198,2,155.61,166.785,"2026-09-16","2026-09-17"],["baseten","glm-4-7","GLM 4.7 (on Baseten)","latency_ttft_ms","ms",465.2,17,287.6,656.4,"2026-09-02","2026-09-18"],["baseten","glm-4-7","GLM 4.7 (on Baseten)","throughput_tps","tokens per second",127.309,17,60.857,164.254,"2026-09-02","2026-09-18"],["baseten","glm-52","GLM-5.2 (on Baseten)","latency_ttft_ms","ms",1295.2,17,382.4,2274.2,"2026-09-02","2026-09-18"],["baseten","glm-52","GLM-5.2 (on Baseten)","throughput_tps","tokens per second",80.236,17,61.321,95.651,"2026-09-02","2026-09-18"],["baseten","glm-53","GLM-5.3 (on Baseten)","latency_ttft_ms","ms",891.8,16,577.2,1081.6,"2026-09-02","2026-09-18"],["baseten","glm-53","GLM-5.3 (on Baseten)","throughput_tps","tokens per second",77.804,16,42.272,87.182,"2026-09-02","2026-09-18"],["baseten","glm-53-flash","GLM-5.3-Flash (on Baseten)","latency_ttft_ms","ms",560.8,17,206.0,736.2,"2026-09-02","2026-09-18"],["baseten","glm-53-flash","GLM-5.3-Flash (on Baseten)","throughput_tps","tokens per second",101.092,17,72.982,167.542,"2026-09-02","2026-09-18"],["baseten","gpt-oss-120b","GPT OSS 120B (on Baseten)","latency_ttft_ms","ms",275.6,17,226.2,458.2,"2026-09-02","2026-09-18"],["baseten","gpt-oss-120b","GPT OSS 120B (on Baseten)","throughput_tps","tokens per second",218.69,17,135.534,282.379,"2026-09-02","2026-09-18"],["baseten","inkling","Inkling (on Baseten)","latency_ttft_ms","ms",237.0,16,177.4,398.4,"2026-09-02","2026-09-18"],["baseten","inkling","Inkling (on Baseten)","throughput_tps","tokens per second",111.097,17,48.786,151.896,"2026-09-02","2026-09-18"],["baseten","inkling-small","Inkling-Small (on Baseten)","latency_ttft_ms","ms",312.3,16,157.4,1703.333,"2026-09-02","2026-09-18"],["baseten","inkling-small","Inkling-Small (on Baseten)","throughput_tps","tokens per second",175.68,16,95.801,230.182,"2026-09-02","2026-09-18"],["baseten","kimi-k26","Kimi K2.6 (on Baseten)","latency_ttft_ms","ms",401.4,17,333.6,584.4,"2026-09-02","2026-09-18"],["baseten","kimi-k26","Kimi K2.6 (on Baseten)","throughput_tps","tokens per second",110.546,17,47.929,149.759,"2026-09-02","2026-09-18"],["baseten","kimi-k27-code","Kimi K2.7 Code (on Baseten)","latency_ttft_ms","ms",319.4,17,239.8,464.0,"2026-09-02","2026-09-18"],["baseten","kimi-k27-code","Kimi K2.7 Code (on Baseten)","throughput_tps","tokens per second",134.057,17,30.522,155.104,"2026-09-02","2026-09-18"],["baseten","kimi-k3","Kimi K3 (on Baseten)","latency_ttft_ms","ms",647.4,16,558.4,1321.4,"2026-09-02","2026-09-18"],["baseten","kimi-k3","Kimi K3 (on Baseten)","throughput_tps","tokens per second",73.105,16,28.845,88.495,"2026-09-02","2026-09-18"],["cerebras","gpt-oss-120b","openai/gpt-oss-120b (on Cerebras)","latency_ttft_ms","ms",188.6,17,127.4,544.0,"2026-09-02","2026-09-18"],["cerebras","gpt-oss-120b","openai/gpt-oss-120b (on Cerebras)","throughput_tps","tokens per second",1076.675,17,515.556,1310.589,"2026-09-02","2026-09-18"],["cerebras","qwen-3-8-27b","Qwen/Qwen3.8-27B (on Cerebras)","latency_ttft_ms","ms",187.2,8,137.6,241.6,"2026-09-11","2026-09-18"],["cerebras","qwen-3-8-27b","Qwen/Qwen3.8-27B (on Cerebras)","throughput_tps","tokens per second",795.57,8,564.323,852.394,"2026-09-11","2026-09-18"],["deepinfra","deepseek-ai--deepseek-r1-0528","deepseek-ai/DeepSeek-R1-0528 (on DeepInfra)","latency_ttft_ms","ms",736.2,17,582.6,1264.2,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-r1-0528","deepseek-ai/DeepSeek-R1-0528 (on DeepInfra)","throughput_tps","tokens per second",18.71,17,13.173,34.69,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3","deepseek-ai/DeepSeek-V3 (on DeepInfra)","latency_ttft_ms","ms",828.6,17,363.2,3690.4,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3","deepseek-ai/DeepSeek-V3 (on DeepInfra)","throughput_tps","tokens per second",15.241,17,3.918,31.434,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3-0324","deepseek-ai/DeepSeek-V3-0324 (on DeepInfra)","latency_ttft_ms","ms",607.0,18,314.8,1081.0,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3-0324","deepseek-ai/DeepSeek-V3-0324 (on DeepInfra)","throughput_tps","tokens per second",39.27,18,13.024,67.805,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3.1","deepseek-ai/DeepSeek-V3.1 (on DeepInfra)","latency_ttft_ms","ms",688.6,17,378.0,1001.6,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3.1","deepseek-ai/DeepSeek-V3.1 (on DeepInfra)","throughput_tps","tokens per second",11.245,17,9.108,20.748,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3.2","deepseek-ai/DeepSeek-V3.2 (on DeepInfra)","latency_ttft_ms","ms",660.9,16,349.8,2453.0,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v3.2","deepseek-ai/DeepSeek-V3.2 (on DeepInfra)","throughput_tps","tokens per second",15.934,16,6.4,34.222,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash","deepseek-ai/DeepSeek-V4-Flash (on DeepInfra)","latency_ttft_ms","ms",572.0,17,397.0,1050.6,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash","deepseek-ai/DeepSeek-V4-Flash (on DeepInfra)","throughput_tps","tokens per second",26.649,17,13.334,44.572,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash-0731","deepseek-ai/DeepSeek-V4-Flash-0731 (on DeepInfra)","latency_ttft_ms","ms",596.0,18,370.2,1253.4,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash-0731","deepseek-ai/DeepSeek-V4-Flash-0731 (on DeepInfra)","throughput_tps","tokens per second",25.692,18,18.12,53.114,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash-vision-exp","deepseek-ai/DeepSeek-V4-Flash-Vision-Exp (on DeepInfra)","latency_ttft_ms","ms",508.3,14,227.6,843.4,"2026-09-05","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-flash-vision-exp","deepseek-ai/DeepSeek-V4-Flash-Vision-Exp (on DeepInfra)","throughput_tps","tokens per second",104.935,14,49.975,141.749,"2026-09-05","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-pro","deepseek-ai/DeepSeek-V4-Pro (on DeepInfra)","latency_ttft_ms","ms",699.2,17,350.0,1421.8,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-pro","deepseek-ai/DeepSeek-V4-Pro (on DeepInfra)","throughput_tps","tokens per second",31.291,17,19.158,56.82,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-pro-0813","deepseek-ai/DeepSeek-V4-Pro-0813 (on DeepInfra)","latency_ttft_ms","ms",354.0,17,224.6,1264.8,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4-pro-0813","deepseek-ai/DeepSeek-V4-Pro-0813 (on DeepInfra)","throughput_tps","tokens per second",62.903,17,38.48,86.283,"2026-09-02","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4.1-flash","deepseek-ai/DeepSeek-V4.1-Flash (on DeepInfra)","latency_ttft_ms","ms",694.4,4,304.8,1771.6,"2026-09-15","2026-09-18"],["deepinfra","deepseek-ai--deepseek-v4.1-flash","deepseek-ai/DeepSeek-V4.1-Flash (on DeepInfra)","throughput_tps","tokens per second",16.398,4,14.269,43.568,"2026-09-15","2026-09-18"],["deepinfra","google--gemma-3-12b-it","google/gemma-3-12b-it (on DeepInfra)","latency_ttft_ms","ms",323.8,17,173.6,963.0,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-3-12b-it","google/gemma-3-12b-it (on DeepInfra)","throughput_tps","tokens per second",42.115,17,12.305,101.055,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-3-27b-it","google/gemma-3-27b-it (on DeepInfra)","latency_ttft_ms","ms",410.8,17,151.2,2082.0,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-3-27b-it","google/gemma-3-27b-it (on DeepInfra)","throughput_tps","tokens per second",19.838,17,14.04,35.574,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-3-4b-it","google/gemma-3-4b-it (on DeepInfra)","latency_ttft_ms","ms",385.8,17,246.2,1187.6,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-3-4b-it","google/gemma-3-4b-it (on DeepInfra)","throughput_tps","tokens per second",32.61,17,8.98,56.612,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-4-26b-a4b-it","google/gemma-4-26B-A4B-it (on DeepInfra)","latency_ttft_ms","ms",388.4,17,225.6,2945.5,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-4-26b-a4b-it","google/gemma-4-26B-A4B-it (on DeepInfra)","throughput_tps","tokens per second",28.176,17,3.507,48.553,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-4-31b-it","google/gemma-4-31B-it (on DeepInfra)","latency_ttft_ms","ms",437.1,18,257.6,1002.8,"2026-09-02","2026-09-18"],["deepinfra","google--gemma-4-31b-it","google/gemma-4-31B-it (on DeepInfra)","throughput_tps","tokens per second",25.109,18,14.854,52.3,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-30b","ibm-granite/granite-4.2-30b (on DeepInfra)","latency_ttft_ms","ms",258.6,15,160.6,354.0,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-30b","ibm-granite/granite-4.2-30b (on DeepInfra)","throughput_tps","tokens per second",85.619,15,43.915,86.444,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-3b","ibm-granite/granite-4.2-3b (on DeepInfra)","latency_ttft_ms","ms",213.8,18,156.0,447.0,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-3b","ibm-granite/granite-4.2-3b (on DeepInfra)","throughput_tps","tokens per second",226.323,18,59.119,237.315,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-8b","ibm-granite/granite-4.2-8b (on DeepInfra)","latency_ttft_ms","ms",252.2,18,173.8,732.2,"2026-09-02","2026-09-18"],["deepinfra","ibm-granite--granite-4.2-8b","ibm-granite/granite-4.2-8b (on DeepInfra)","throughput_tps","tokens per second",127.356,18,29.798,154.728,"2026-09-02","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash","inclusionAI/Ling-3.0-flash (on DeepInfra)","latency_ttft_ms","ms",541.2,16,341.8,1128.6,"2026-09-02","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash","inclusionAI/Ling-3.0-flash (on DeepInfra)","throughput_tps","tokens per second",40.705,16,26.442,105.337,"2026-09-02","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash-fin","inclusionAI/Ling-3.0-flash-Fin (on DeepInfra)","latency_ttft_ms","ms",435.0,14,390.6,542.6,"2026-09-05","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash-fin","inclusionAI/Ling-3.0-flash-Fin (on DeepInfra)","throughput_tps","tokens per second",81.235,14,19.427,194.416,"2026-09-05","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash-vl","inclusionAI/Ling-3.0-flash-VL (on DeepInfra)","latency_ttft_ms","ms",508.2,4,299.0,2272.8,"2026-09-15","2026-09-18"],["deepinfra","inclusionai--ling-3.0-flash-vl","inclusionAI/Ling-3.0-flash-VL (on DeepInfra)","throughput_tps","tokens per second",44.468,4,27.948,104.915,"2026-09-15","2026-09-18"],["deepinfra","meta-llama--llama-4-scout-17b-16e-instruct","meta-llama/Llama-4-Scout-17B-16E-Instruct (on DeepInfra)","latency_ttft_ms","ms",531.2,17,244.0,1063.8,"2026-09-02","2026-09-18"],["deepinfra","meta-llama--llama-4-scout-17b-16e-instruct","meta-llama/Llama-4-Scout-17B-16E-Instruct (on DeepInfra)","throughput_tps","tokens per second",32.895,17,12.355,70.783,"2026-09-02","2026-09-18"],["deepinfra","meta-llama--llama-guard-4-12b","meta-llama/Llama-Guard-4-12B (on DeepInfra)","latency_ttft_ms","ms",364.4,17,212.6,3493.0,"2026-09-02","2026-09-18"],["deepinfra","meta-llama--llama-guard-4-12b","meta-llama/Llama-Guard-4-12B (on DeepInfra)","throughput_tps","tokens per second",7.825,17,0.46,11.446,"2026-09-02","2026-09-18"],["deepinfra","meta-models--muse-glimmer-30b","meta-models/Muse-Glimmer-30B (on DeepInfra)","latency_ttft_ms","ms",246.8,17,179.4,1503.8,"2026-09-02","2026-09-18"],["deepinfra","meta-models--muse-glimmer-30b","meta-models/Muse-Glimmer-30B (on DeepInfra)","throughput_tps","tokens per second",107.543,17,59.044,122.861,"2026-09-02","2026-09-18"],["deepinfra","microsoft--phi-4","microsoft/phi-4 (on DeepInfra)","latency_ttft_ms","ms",321.2,17,195.2,501.2,"2026-09-02","2026-09-18"],["deepinfra","microsoft--phi-4","microsoft/phi-4 (on DeepInfra)","throughput_tps","tokens per second",67.772,17,42.53,78.916,"2026-09-02","2026-09-18"],["deepinfra","minimaxai--minimax-m2.7","MiniMaxAI/MiniMax-M2.7 (on DeepInfra)","latency_ttft_ms","ms",682.4,17,478.8,2851.2,"2026-09-02","2026-09-18"],["deepinfra","minimaxai--minimax-m2.7","MiniMaxAI/MiniMax-M2.7 (on DeepInfra)","throughput_tps","tokens per second",41.192,17,14.498,112.846,"2026-09-02","2026-09-18"],["deepinfra","minimaxai--minimax-m3","MiniMaxAI/MiniMax-M3 (on DeepInfra)","latency_ttft_ms","ms",637.0,18,385.4,1876.6,"2026-09-02","2026-09-18"],["deepinfra","minimaxai--minimax-m3","MiniMaxAI/MiniMax-M3 (on DeepInfra)","throughput_tps","tokens per second",27.01,18,12.949,52.847,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k2.6","moonshotai/Kimi-K2.6 (on DeepInfra)","latency_ttft_ms","ms",812.8,17,603.4,1664.4,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k2.6","moonshotai/Kimi-K2.6 (on DeepInfra)","throughput_tps","tokens per second",18.489,17,9.132,38.918,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k2.7-code","moonshotai/Kimi-K2.7-Code (on DeepInfra)","latency_ttft_ms","ms",892.8,17,574.8,1753.4,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k2.7-code","moonshotai/Kimi-K2.7-Code (on DeepInfra)","throughput_tps","tokens per second",31.359,17,14.056,72.086,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k3","moonshotai/Kimi-K3 (on DeepInfra)","latency_ttft_ms","ms",814.0,16,356.4,3618.0,"2026-09-02","2026-09-18"],["deepinfra","moonshotai--kimi-k3","moonshotai/Kimi-K3 (on DeepInfra)","throughput_tps","tokens per second",20.45,16,2.642,69.438,"2026-09-02","2026-09-18"],["deepinfra","nousresearch--hermes-3-llama-3.1-70b","NousResearch/Hermes-3-Llama-3.1-70B (on DeepInfra)","latency_ttft_ms","ms",209.9,18,163.0,517.4,"2026-09-02","2026-09-18"],["deepinfra","nousresearch--hermes-3-llama-3.1-70b","NousResearch/Hermes-3-Llama-3.1-70B (on DeepInfra)","throughput_tps","tokens per second",32.358,18,10.274,38.678,"2026-09-02","2026-09-18"],["deepinfra","openai--gpt-oss-120b","openai/gpt-oss-120b (on DeepInfra)","latency_ttft_ms","ms",375.5,18,214.2,1738.8,"2026-09-02","2026-09-18"],["deepinfra","openai--gpt-oss-120b","openai/gpt-oss-120b (on DeepInfra)","throughput_tps","tokens per second",45.032,18,21.541,77.777,"2026-09-02","2026-09-18"],["deepinfra","openai--gpt-oss-20b","openai/gpt-oss-20b (on DeepInfra)","latency_ttft_ms","ms",289.8,17,220.0,2101.8,"2026-09-02","2026-09-18"],["deepinfra","openai--gpt-oss-20b","openai/gpt-oss-20b (on DeepInfra)","throughput_tps","tokens per second",98.393,17,60.786,140.735,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen2.5-72b-instruct","Qwen/Qwen2.5-72B-Instruct (on DeepInfra)","latency_ttft_ms","ms",379.8,17,216.0,1066.6,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen2.5-72b-instruct","Qwen/Qwen2.5-72B-Instruct (on DeepInfra)","throughput_tps","tokens per second",31.194,17,15.693,51.828,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-14b","Qwen/Qwen3-14B (on DeepInfra)","latency_ttft_ms","ms",220.0,17,179.6,538.0,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-14b","Qwen/Qwen3-14B (on DeepInfra)","throughput_tps","tokens per second",68.179,17,32.292,83.596,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-235b-a22b-instruct-2507","Qwen/Qwen3-235B-A22B-Instruct-2507 (on DeepInfra)","latency_ttft_ms","ms",536.6,17,246.2,4427.8,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-235b-a22b-instruct-2507","Qwen/Qwen3-235B-A22B-Instruct-2507 (on DeepInfra)","throughput_tps","tokens per second",14.376,17,3.985,57.943,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-30b-a3b","Qwen/Qwen3-30B-A3B (on DeepInfra)","latency_ttft_ms","ms",228.2,17,180.0,498.8,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-30b-a3b","Qwen/Qwen3-30B-A3B (on DeepInfra)","throughput_tps","tokens per second",119.31,17,53.63,142.154,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-32b","Qwen/Qwen3-32B (on DeepInfra)","latency_ttft_ms","ms",281.0,17,190.4,1540.4,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-32b","Qwen/Qwen3-32B (on DeepInfra)","throughput_tps","tokens per second",29.03,17,19.758,47.434,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-next-80b-a3b-instruct","Qwen/Qwen3-Next-80B-A3B-Instruct (on DeepInfra)","latency_ttft_ms","ms",487.7,18,340.4,729.8,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-next-80b-a3b-instruct","Qwen/Qwen3-Next-80B-A3B-Instruct (on DeepInfra)","throughput_tps","tokens per second",87.178,18,48.506,145.946,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-vl-235b-a22b-instruct","Qwen/Qwen3-VL-235B-A22B-Instruct (on DeepInfra)","latency_ttft_ms","ms",648.1,18,217.6,2107.2,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-vl-235b-a22b-instruct","Qwen/Qwen3-VL-235B-A22B-Instruct (on DeepInfra)","throughput_tps","tokens per second",18.395,18,8.449,32.466,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3-vl-30b-a3b-instruct","Qwen/Qwen3-VL-30B-A3B-Instruct (on DeepInfra)","latency_ttft_ms","ms",340.1,16,153.6,939.6,"2026-09-02","2026-09-17"],["deepinfra","qwen--qwen3-vl-30b-a3b-instruct","Qwen/Qwen3-VL-30B-A3B-Instruct (on DeepInfra)","throughput_tps","tokens per second",28.291,17,13.196,42.762,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-122b-a10b","Qwen/Qwen3.5-122B-A10B (on DeepInfra)","latency_ttft_ms","ms",433.2,18,333.2,1998.6,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-122b-a10b","Qwen/Qwen3.5-122B-A10B (on DeepInfra)","throughput_tps","tokens per second",109.203,18,55.205,126.585,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-27b","Qwen/Qwen3.5-27B (on DeepInfra)","latency_ttft_ms","ms",290.4,17,182.5,2039.0,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-27b","Qwen/Qwen3.5-27B (on DeepInfra)","throughput_tps","tokens per second",73.29,17,57.622,84.86,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-35b-a3b","Qwen/Qwen3.5-35B-A3B (on DeepInfra)","latency_ttft_ms","ms",290.6,18,205.6,726.6,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-35b-a3b","Qwen/Qwen3.5-35B-A3B (on DeepInfra)","throughput_tps","tokens per second",116.119,18,31.131,146.55,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-397b-a17b","Qwen/Qwen3.5-397B-A17B (on DeepInfra)","latency_ttft_ms","ms",385.0,17,210.8,659.2,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-397b-a17b","Qwen/Qwen3.5-397B-A17B (on DeepInfra)","throughput_tps","tokens per second",44.877,17,23.36,81.813,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-9b","Qwen/Qwen3.5-9B (on DeepInfra)","latency_ttft_ms","ms",365.4,17,253.8,697.6,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.5-9b","Qwen/Qwen3.5-9B (on DeepInfra)","throughput_tps","tokens per second",31.735,17,9.031,83.351,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.6-27b","Qwen/Qwen3.6-27B (on DeepInfra)","latency_ttft_ms","ms",242.4,17,189.8,665.2,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.6-27b","Qwen/Qwen3.6-27B (on DeepInfra)","throughput_tps","tokens per second",65.378,17,21.072,83.376,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.6-35b-a3b","Qwen/Qwen3.6-35B-A3B (on DeepInfra)","latency_ttft_ms","ms",404.8,16,240.4,2656.667,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.6-35b-a3b","Qwen/Qwen3.6-35B-A3B (on DeepInfra)","throughput_tps","tokens per second",46.781,16,18.763,135.376,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.8-2.4t-a95b","Qwen/Qwen3.8-2.4T-A95B (on DeepInfra)","latency_ttft_ms","ms",476.0,17,377.8,1284.8,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.8-2.4t-a95b","Qwen/Qwen3.8-2.4T-A95B (on DeepInfra)","throughput_tps","tokens per second",73.727,17,35.399,91.311,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.8-27b","Qwen/Qwen3.8-27B (on DeepInfra)","latency_ttft_ms","ms",348.0,17,274.6,2641.2,"2026-09-02","2026-09-18"],["deepinfra","qwen--qwen3.8-27b","Qwen/Qwen3.8-27B (on DeepInfra)","throughput_tps","tokens per second",32.349,17,11.213,56.501,"2026-09-02","2026-09-18"],["deepinfra","stepfun-ai--step-3.7-flash","stepfun-ai/Step-3.7-Flash (on DeepInfra)","latency_ttft_ms","ms",273.4,17,178.0,503.0,"2026-09-02","2026-09-18"],["deepinfra","stepfun-ai--step-3.7-flash","stepfun-ai/Step-3.7-Flash (on DeepInfra)","throughput_tps","tokens per second",119.679,17,82.949,141.017,"2026-09-02","2026-09-18"],["deepinfra","tencent--hy3","tencent/Hy3 (on DeepInfra)","latency_ttft_ms","ms",382.2,17,179.0,486.6,"2026-09-02","2026-09-18"],["deepinfra","tencent--hy3","tencent/Hy3 (on DeepInfra)","throughput_tps","tokens per second",60.399,17,28.647,111.014,"2026-09-02","2026-09-18"],["deepinfra","thinkingmachines--inkling","thinkingmachines/Inkling (on DeepInfra)","latency_ttft_ms","ms",565.4,17,512.8,883.8,"2026-09-02","2026-09-18"],["deepinfra","thinkingmachines--inkling","thinkingmachines/Inkling (on DeepInfra)","throughput_tps","tokens per second",98.939,17,14.064,118.131,"2026-09-02","2026-09-18"],["deepinfra","thinkingmachines--inkling-small","thinkingmachines/Inkling-Small (on DeepInfra)","latency_ttft_ms","ms",356.7,18,273.6,1225.4,"2026-09-02","2026-09-18"],["deepinfra","thinkingmachines--inkling-small","thinkingmachines/Inkling-Small (on DeepInfra)","throughput_tps","tokens per second",128.352,18,15.44,142.763,"2026-09-02","2026-09-18"],["deepinfra","xiaomimimo--mimo-v2.5","XiaomiMiMo/MiMo-V2.5 (on DeepInfra)","latency_ttft_ms","ms",1109.2,18,385.4,2307.8,"2026-09-02","2026-09-18"],["deepinfra","xiaomimimo--mimo-v2.5","XiaomiMiMo/MiMo-V2.5 (on DeepInfra)","throughput_tps","tokens per second",14.773,18,3.711,51.777,"2026-09-02","2026-09-18"],["deepinfra","xiaomimimo--mimo-v2.5-pro","XiaomiMiMo/MiMo-V2.5-Pro (on DeepInfra)","latency_ttft_ms","ms",364.9,18,266.8,1185.4,"2026-09-02","2026-09-18"],["deepinfra","xiaomimimo--mimo-v2.5-pro","XiaomiMiMo/MiMo-V2.5-Pro (on DeepInfra)","throughput_tps","tokens per second",34.356,18,6.687,46.633,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.6","zai-org/GLM-4.6 (on DeepInfra)","latency_ttft_ms","ms",535.4,17,354.8,3016.8,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.6","zai-org/GLM-4.6 (on DeepInfra)","throughput_tps","tokens per second",28.241,17,19.334,72.174,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.7","zai-org/GLM-4.7 (on DeepInfra)","latency_ttft_ms","ms",751.1,16,545.0,2027.8,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.7","zai-org/GLM-4.7 (on DeepInfra)","throughput_tps","tokens per second",15.089,16,11.13,29.101,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.7-flash","zai-org/GLM-4.7-Flash (on DeepInfra)","latency_ttft_ms","ms",723.9,18,316.0,2322.8,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-4.7-flash","zai-org/GLM-4.7-Flash (on DeepInfra)","throughput_tps","tokens per second",25.751,18,11.8,104.326,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5","zai-org/GLM-5 (on DeepInfra)","latency_ttft_ms","ms",698.8,17,315.0,2016.8,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5","zai-org/GLM-5 (on DeepInfra)","throughput_tps","tokens per second",29.087,17,21.163,93.107,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5.1","zai-org/GLM-5.1 (on DeepInfra)","latency_ttft_ms","ms",1021.6,16,804.6,4595.5,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5.1","zai-org/GLM-5.1 (on DeepInfra)","throughput_tps","tokens per second",21.76,16,7.996,53.025,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5.2","zai-org/GLM-5.2 (on DeepInfra)","latency_ttft_ms","ms",476.2,17,235.6,1664.0,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5.2","zai-org/GLM-5.2 (on DeepInfra)","throughput_tps","tokens per second",49.776,17,33.24,92.844,"2026-09-02","2026-09-18"],["deepinfra","zai-org--glm-5.3","zai-org/GLM-5.3 (on DeepInfra)","latency_ttft_ms","ms",508.2,13,352.6,968.2,"2026-09-05","2026-09-18"],["deepinfra","zai-org--glm-5.3","zai-org/GLM-5.3 (on DeepInfra)","throughput_tps","tokens per second",46.883,13,25.986,83.304,"2026-09-05","2026-09-18"],["deepinfra","zai-org--glm-5.3-flash","zai-org/GLM-5.3-Flash (on DeepInfra)","latency_ttft_ms","ms",673.3,14,390.4,1877.8,"2026-09-05","2026-09-18"],["deepinfra","zai-org--glm-5.3-flash","zai-org/GLM-5.3-Flash (on DeepInfra)","throughput_tps","tokens per second",22.523,14,10.633,47.956,"2026-09-05","2026-09-18"],["fireworks_ai","deepseek-v4-flash-0731","DeepSeek-V4-Flash-0731 (on Fireworks AI)","latency_ttft_ms","ms",1023.0,17,657.6,2154.5,"2026-09-02","2026-09-18"],["fireworks_ai","deepseek-v4-flash-0731","DeepSeek-V4-Flash-0731 (on Fireworks AI)","throughput_tps","tokens per second",77.563,17,60.297,97.451,"2026-09-02","2026-09-18"],["fireworks_ai","deepseek-v4-flash-vision-exp","DeepSeek-V4-Flash-Vision-Exp (on Fireworks AI)","latency_ttft_ms","ms",589.1,14,409.2,1018.8,"2026-09-05","2026-09-18"],["fireworks_ai","deepseek-v4-flash-vision-exp","DeepSeek-V4-Flash-Vision-Exp (on Fireworks AI)","throughput_tps","tokens per second",95.193,14,68.628,137.984,"2026-09-05","2026-09-18"],["fireworks_ai","deepseek-v4-pro-0813","DeepSeek-V4-Pro-0813 (on Fireworks AI)","latency_ttft_ms","ms",942.0,17,561.0,1749.4,"2026-09-02","2026-09-18"],["fireworks_ai","deepseek-v4-pro-0813","DeepSeek-V4-Pro-0813 (on Fireworks AI)","throughput_tps","tokens per second",53.086,17,48.503,65.165,"2026-09-02","2026-09-18"],["fireworks_ai","deepseek-v4p1-flash","DeepSeek V4.1 Flash (on Fireworks AI)","latency_ttft_ms","ms",1135.8,7,657.2,1941.4,"2026-09-12","2026-09-18"],["fireworks_ai","deepseek-v4p1-flash","DeepSeek V4.1 Flash (on Fireworks AI)","throughput_tps","tokens per second",59.848,7,45.144,77.774,"2026-09-12","2026-09-18"],["fireworks_ai","glm-5p2","GLM 5.2 (on Fireworks AI)","latency_ttft_ms","ms",594.6,16,454.2,1225.6,"2026-09-02","2026-09-18"],["fireworks_ai","glm-5p2","GLM 5.2 (on Fireworks AI)","throughput_tps","tokens per second",67.729,17,46.586,73.193,"2026-09-02","2026-09-18"],["fireworks_ai","glm-5p3","GLM-5.3 (on Fireworks AI)","latency_ttft_ms","ms",567.4,17,402.4,1590.0,"2026-09-02","2026-09-18"],["fireworks_ai","glm-5p3","GLM-5.3 (on Fireworks AI)","throughput_tps","tokens per second",59.536,17,45.434,74.161,"2026-09-02","2026-09-18"],["fireworks_ai","glm-5p3-flash","GLM 5.3 Flash (on Fireworks AI)","latency_ttft_ms","ms",652.6,17,374.4,2569.4,"2026-09-02","2026-09-18"],["fireworks_ai","glm-5p3-flash","GLM 5.3 Flash (on Fireworks AI)","throughput_tps","tokens per second",76.251,17,39.924,101.045,"2026-09-02","2026-09-18"],["fireworks_ai","inkling","Inkling (on Fireworks AI)","latency_ttft_ms","ms",385.2,17,314.8,570.6,"2026-09-02","2026-09-18"],["fireworks_ai","inkling","Inkling (on Fireworks AI)","throughput_tps","tokens per second",93.461,17,33.912,106.653,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k2p6","Kimi K2.6 (on Fireworks AI)","latency_ttft_ms","ms",419.4,17,276.0,1567.6,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k2p6","Kimi K2.6 (on Fireworks AI)","throughput_tps","tokens per second",60.297,17,35.322,89.231,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k2p7-code","Kimi K2.7 Code (on Fireworks AI)","latency_ttft_ms","ms",401.8,17,282.4,1553.0,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k2p7-code","Kimi K2.7 Code (on Fireworks AI)","throughput_tps","tokens per second",87.79,17,54.497,106.515,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k3","Kimi K3 (on Fireworks AI)","latency_ttft_ms","ms",1493.2,16,714.4,2979.0,"2026-09-02","2026-09-18"],["fireworks_ai","kimi-k3","Kimi K3 (on Fireworks AI)","throughput_tps","tokens per second",42.423,16,35.351,57.226,"2026-09-02","2026-09-18"],["fireworks_ai","minimax-m3","Minimax M3 (on Fireworks AI)","latency_ttft_ms","ms",724.6,17,459.6,2155.4,"2026-09-02","2026-09-18"],["fireworks_ai","minimax-m3","Minimax M3 (on Fireworks AI)","throughput_tps","tokens per second",94.275,17,70.445,122.792,"2026-09-02","2026-09-18"],["fireworks_ai","muse-glimmer-30b","Muse Glimmer 30B (on Fireworks AI)","latency_ttft_ms","ms",102.0,17,89.0,317.6,"2026-09-02","2026-09-18"],["fireworks_ai","muse-glimmer-30b","Muse Glimmer 30B (on Fireworks AI)","throughput_tps","tokens per second",60.695,17,43.641,63.389,"2026-09-02","2026-09-18"],["groq","openai--gpt-oss-120b","GPT OSS 120B (on Groq)","latency_ttft_ms","ms",273.8,17,130.8,461.8,"2026-09-02","2026-09-18"],["groq","openai--gpt-oss-120b","GPT OSS 120B (on Groq)","throughput_tps","tokens per second",433.376,17,401.966,444.203,"2026-09-02","2026-09-18"],["groq","openai--gpt-oss-20b","GPT OSS 20B (on Groq)","latency_ttft_ms","ms",247.6,17,174.2,456.2,"2026-09-02","2026-09-18"],["groq","openai--gpt-oss-20b","GPT OSS 20B (on Groq)","throughput_tps","tokens per second",725.374,17,658.086,777.032,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-r1","deepseek/deepseek-r1 (on Novita AI)","latency_ttft_ms","ms",1032.6,16,754.5,1650.6,"2026-09-03","2026-09-18"],["novita","deepseek--deepseek-r1","deepseek/deepseek-r1 (on Novita AI)","throughput_tps","tokens per second",19.442,16,12.532,24.314,"2026-09-03","2026-09-18"],["novita","deepseek--deepseek-r1-0528","deepseek/deepseek-r1-0528 (on Novita AI)","latency_ttft_ms","ms",1040.6,17,793.2,1632.8,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-r1-0528","deepseek/deepseek-r1-0528 (on Novita AI)","throughput_tps","tokens per second",20.077,17,12.482,25.913,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-r1-distill-llama-70b","deepseek/deepseek-r1-distill-llama-70b (on Novita AI)","latency_ttft_ms","ms",942.8,17,782.8,2229.2,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-r1-distill-llama-70b","deepseek/deepseek-r1-distill-llama-70b (on Novita AI)","throughput_tps","tokens per second",20.779,17,10.128,30.905,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.1","deepseek/deepseek-v3.1 (on Novita AI)","latency_ttft_ms","ms",1597.2,17,1165.8,1907.0,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.1","deepseek/deepseek-v3.1 (on Novita AI)","throughput_tps","tokens per second",31.429,17,28.886,36.262,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.1-terminus","deepseek/deepseek-v3.1-terminus (on Novita AI)","latency_ttft_ms","ms",1480.4,17,1270.0,1668.6,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.1-terminus","deepseek/deepseek-v3.1-terminus (on Novita AI)","throughput_tps","tokens per second",27.195,17,24.011,35.326,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.2","deepseek/deepseek-v3.2 (on Novita AI)","latency_ttft_ms","ms",1305.2,17,964.8,1704.0,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.2","deepseek/deepseek-v3.2 (on Novita AI)","throughput_tps","tokens per second",36.48,17,25.593,42.061,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.2-exp","deepseek/deepseek-v3.2-exp (on Novita AI)","latency_ttft_ms","ms",1245.4,17,1071.6,1526.2,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v3.2-exp","deepseek/deepseek-v3.2-exp (on Novita AI)","throughput_tps","tokens per second",30.898,17,26.378,36.198,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-flash","deepseek/deepseek-v4-flash (on Novita AI)","latency_ttft_ms","ms",1225.8,17,1088.2,1665.6,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-flash","deepseek/deepseek-v4-flash (on Novita AI)","throughput_tps","tokens per second",92.875,17,78.521,102.041,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-flash-0731","deepseek/deepseek-v4-flash-0731 (on Novita AI)","latency_ttft_ms","ms",769.0,17,402.2,1166.6,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-flash-0731","deepseek/deepseek-v4-flash-0731 (on Novita AI)","throughput_tps","tokens per second",88.055,17,77.431,136.899,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-flash-vision-exp","deepseek/deepseek-v4-flash-vision-exp (on Novita AI)","latency_ttft_ms","ms",512.6,15,443.4,597.2,"2026-09-04","2026-09-18"],["novita","deepseek--deepseek-v4-flash-vision-exp","deepseek/deepseek-v4-flash-vision-exp (on Novita AI)","throughput_tps","tokens per second",123.748,15,91.735,134.293,"2026-09-04","2026-09-18"],["novita","deepseek--deepseek-v4-pro","deepseek/deepseek-v4-pro (on Novita AI)","latency_ttft_ms","ms",645.4,17,494.2,1252.4,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-pro","deepseek/deepseek-v4-pro (on Novita AI)","throughput_tps","tokens per second",39.895,17,30.391,78.978,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-pro-0813","deepseek/deepseek-v4-pro-0813 (on Novita AI)","latency_ttft_ms","ms",614.4,17,457.6,1228.0,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4-pro-0813","deepseek/deepseek-v4-pro-0813 (on Novita AI)","throughput_tps","tokens per second",41.132,17,31.238,63.894,"2026-09-02","2026-09-18"],["novita","deepseek--deepseek-v4.1-flash","deepseek/deepseek-v4.1-flash (on Novita AI)","latency_ttft_ms","ms",564.3,8,414.2,917.0,"2026-09-11","2026-09-18"],["novita","deepseek--deepseek-v4.1-flash","deepseek/deepseek-v4.1-flash (on Novita AI)","throughput_tps","tokens per second",123.077,8,89.244,131.015,"2026-09-11","2026-09-18"],["novita","google--gemma-4-26b-a4b-it","google/gemma-4-26b-a4b-it (on Novita AI)","latency_ttft_ms","ms",762.0,17,428.2,1011.0,"2026-09-02","2026-09-18"],["novita","google--gemma-4-26b-a4b-it","google/gemma-4-26b-a4b-it (on Novita AI)","throughput_tps","tokens per second",60.054,17,23.629,146.62,"2026-09-02","2026-09-18"],["novita","google--gemma-4-31b-it","google/gemma-4-31b-it (on Novita AI)","latency_ttft_ms","ms",564.2,17,457.0,1452.8,"2026-09-02","2026-09-18"],["novita","google--gemma-4-31b-it","google/gemma-4-31b-it (on Novita AI)","throughput_tps","tokens per second",59.896,17,19.908,125.771,"2026-09-02","2026-09-18"],["novita","inclusionai--ling-3.0-flash","inclusionai/ling-3.0-flash (on Novita AI)","latency_ttft_ms","ms",718.2,17,626.4,871.4,"2026-09-02","2026-09-18"],["novita","inclusionai--ling-3.0-flash","inclusionai/ling-3.0-flash (on Novita AI)","throughput_tps","tokens per second",233.371,17,100.26,286.873,"2026-09-02","2026-09-18"],["novita","inclusionai--ling-3.0-flash-fin","inclusionai/ling-3.0-flash-fin (on Novita AI)","latency_ttft_ms","ms",669.0,10,542.8,760.8,"2026-09-09","2026-09-18"],["novita","inclusionai--ling-3.0-flash-fin","inclusionai/ling-3.0-flash-fin (on Novita AI)","throughput_tps","tokens per second",263.519,10,107.602,288.153,"2026-09-09","2026-09-18"],["novita","inclusionai--ling-3.0-flash-vl","inclusionai/ling-3.0-flash-vl (on Novita AI)","latency_ttft_ms","ms",728.2,9,652.6,784.8,"2026-09-10","2026-09-18"],["novita","inclusionai--ling-3.0-flash-vl","inclusionai/ling-3.0-flash-vl (on Novita AI)","throughput_tps","tokens per second",139.063,9,109.363,142.152,"2026-09-10","2026-09-18"],["novita","meta-llama--llama-3.1-8b-instruct","meta-llama/llama-3.1-8b-instruct (on Novita AI)","latency_ttft_ms","ms",535.6,17,409.2,2062.5,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-3.1-8b-instruct","meta-llama/llama-3.1-8b-instruct (on Novita AI)","throughput_tps","tokens per second",127.684,17,100.594,190.959,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-3.3-70b-instruct","meta-llama/llama-3.3-70b-instruct (on Novita AI)","latency_ttft_ms","ms",526.5,16,408.0,688.2,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-3.3-70b-instruct","meta-llama/llama-3.3-70b-instruct (on Novita AI)","throughput_tps","tokens per second",46.668,16,36.31,50.906,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-4-maverick-17b-128e-instruct-fp8","meta-llama/llama-4-maverick-17b-128e-instruct-fp8 (on Novita AI)","latency_ttft_ms","ms",400.2,17,342.8,524.2,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-4-maverick-17b-128e-instruct-fp8","meta-llama/llama-4-maverick-17b-128e-instruct-fp8 (on Novita AI)","throughput_tps","tokens per second",80.378,17,52.709,102.716,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-4-scout-17b-16e-instruct","meta-llama/llama-4-scout-17b-16e-instruct (on Novita AI)","latency_ttft_ms","ms",576.4,17,327.8,1796.2,"2026-09-02","2026-09-18"],["novita","meta-llama--llama-4-scout-17b-16e-instruct","meta-llama/llama-4-scout-17b-16e-instruct (on Novita AI)","throughput_tps","tokens per second",34.419,17,15.963,61.131,"2026-09-02","2026-09-18"],["novita","microsoft--wizardlm-2-8x22b","microsoft/wizardlm-2-8x22b (on Novita AI)","latency_ttft_ms","ms",713.6,17,386.0,2411.6,"2026-09-02","2026-09-18"],["novita","microsoft--wizardlm-2-8x22b","microsoft/wizardlm-2-8x22b (on Novita AI)","throughput_tps","tokens per second",18.124,17,11.218,66.564,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2","minimax/minimax-m2 (on Novita AI)","latency_ttft_ms","ms",1324.4,17,1060.0,1608.8,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2","minimax/minimax-m2 (on Novita AI)","throughput_tps","tokens per second",64.708,17,56.189,73.131,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.1","minimax/minimax-m2.1 (on Novita AI)","latency_ttft_ms","ms",1348.8,17,1132.6,1927.75,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.1","minimax/minimax-m2.1 (on Novita AI)","throughput_tps","tokens per second",58.683,17,46.786,69.096,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.5","minimax/minimax-m2.5 (on Novita AI)","latency_ttft_ms","ms",1012.0,17,874.6,1536.2,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.5","minimax/minimax-m2.5 (on Novita AI)","throughput_tps","tokens per second",60.652,17,52.374,69.87,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.7","minimax/minimax-m2.7 (on Novita AI)","latency_ttft_ms","ms",1061.4,17,862.2,1689.2,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m2.7","minimax/minimax-m2.7 (on Novita AI)","throughput_tps","tokens per second",47.039,17,28.978,64.984,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m3","minimax/minimax-m3 (on Novita AI)","latency_ttft_ms","ms",957.2,17,675.2,1743.6,"2026-09-02","2026-09-18"],["novita","minimax--minimax-m3","minimax/minimax-m3 (on Novita AI)","throughput_tps","tokens per second",75.42,17,37.538,155.506,"2026-09-02","2026-09-18"],["novita","minimaxai--minimax-m1-80k","minimaxai/minimax-m1-80k (on Novita AI)","latency_ttft_ms","ms",1069.2,17,949.6,2382.8,"2026-09-02","2026-09-18"],["novita","minimaxai--minimax-m1-80k","minimaxai/minimax-m1-80k (on Novita AI)","throughput_tps","tokens per second",55.445,17,39.549,59.329,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2-instruct","moonshotai/kimi-k2-instruct (on Novita AI)","latency_ttft_ms","ms",995.4,17,735.8,1551.0,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2-instruct","moonshotai/kimi-k2-instruct (on Novita AI)","throughput_tps","tokens per second",37.471,17,33.787,41.024,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.5","moonshotai/kimi-k2.5 (on Novita AI)","latency_ttft_ms","ms",1157.0,17,691.4,2212.667,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.5","moonshotai/kimi-k2.5 (on Novita AI)","throughput_tps","tokens per second",38.169,17,34.096,44.14,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.6","moonshotai/kimi-k2.6 (on Novita AI)","latency_ttft_ms","ms",1250.2,17,1000.4,1706.6,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.6","moonshotai/kimi-k2.6 (on Novita AI)","throughput_tps","tokens per second",58.035,17,32.204,82.623,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.7-code","moonshotai/kimi-k2.7-code (on Novita AI)","latency_ttft_ms","ms",1331.4,16,950.2,1877.8,"2026-09-02","2026-09-18"],["novita","moonshotai--kimi-k2.7-code","moonshotai/kimi-k2.7-code (on Novita AI)","throughput_tps","tokens per second",56.797,16,22.642,81.608,"2026-09-02","2026-09-18"],["novita","openai--gpt-oss-120b","openai/gpt-oss-120b (on Novita AI)","latency_ttft_ms","ms",498.8,17,378.4,2819.0,"2026-09-02","2026-09-18"],["novita","openai--gpt-oss-120b","openai/gpt-oss-120b (on Novita AI)","throughput_tps","tokens per second",99.439,17,19.727,212.662,"2026-09-02","2026-09-18"],["novita","openai--gpt-oss-20b","openai/gpt-oss-20b (on Novita AI)","latency_ttft_ms","ms",448.6,17,299.2,628.8,"2026-09-02","2026-09-18"],["novita","openai--gpt-oss-20b","openai/gpt-oss-20b (on Novita AI)","throughput_tps","tokens per second",157.843,17,68.667,243.09,"2026-09-02","2026-09-18"],["novita","qwen--qwen-2.5-72b-instruct","qwen/qwen-2.5-72b-instruct (on Novita AI)","latency_ttft_ms","ms",572.4,17,390.8,2249.4,"2026-09-02","2026-09-18"],["novita","qwen--qwen-2.5-72b-instruct","qwen/qwen-2.5-72b-instruct (on Novita AI)","throughput_tps","tokens per second",28.842,17,10.871,38.376,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-235b-a22b-instruct-2507","qwen/qwen3-235b-a22b-instruct-2507 (on Novita AI)","latency_ttft_ms","ms",654.8,17,519.4,942.4,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-235b-a22b-instruct-2507","qwen/qwen3-235b-a22b-instruct-2507 (on Novita AI)","throughput_tps","tokens per second",43.605,17,37.382,56.589,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-235b-a22b-thinking-2507","qwen/qwen3-235b-a22b-thinking-2507 (on Novita AI)","latency_ttft_ms","ms",875.8,17,617.6,998.8,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-235b-a22b-thinking-2507","qwen/qwen3-235b-a22b-thinking-2507 (on Novita AI)","throughput_tps","tokens per second",60.352,17,42.414,70.664,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-coder-480b-a35b-instruct","qwen/qwen3-coder-480b-a35b-instruct (on Novita AI)","latency_ttft_ms","ms",944.4,17,645.2,1599.8,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-coder-480b-a35b-instruct","qwen/qwen3-coder-480b-a35b-instruct (on Novita AI)","throughput_tps","tokens per second",51.356,17,46.385,66.075,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-coder-next","qwen/qwen3-coder-next (on Novita AI)","latency_ttft_ms","ms",1315.6,17,838.4,2131.6,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-coder-next","qwen/qwen3-coder-next (on Novita AI)","throughput_tps","tokens per second",119.055,17,105.957,128.834,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-next-80b-a3b-instruct","qwen/qwen3-next-80b-a3b-instruct (on Novita AI)","latency_ttft_ms","ms",775.0,17,560.8,929.6,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-next-80b-a3b-instruct","qwen/qwen3-next-80b-a3b-instruct (on Novita AI)","throughput_tps","tokens per second",141.11,17,99.035,147.323,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-235b-a22b-instruct","qwen/qwen3-vl-235b-a22b-instruct (on Novita AI)","latency_ttft_ms","ms",673.2,17,576.0,1062.2,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-235b-a22b-instruct","qwen/qwen3-vl-235b-a22b-instruct (on Novita AI)","throughput_tps","tokens per second",43.541,17,24.111,58.439,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-235b-a22b-thinking","qwen/qwen3-vl-235b-a22b-thinking (on Novita AI)","latency_ttft_ms","ms",983.6,17,785.4,1130.0,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-235b-a22b-thinking","qwen/qwen3-vl-235b-a22b-thinking (on Novita AI)","throughput_tps","tokens per second",51.659,17,38.101,59.531,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-30b-a3b-instruct","qwen/qwen3-vl-30b-a3b-instruct (on Novita AI)","latency_ttft_ms","ms",618.6,17,507.0,884.8,"2026-09-02","2026-09-18"],["novita","qwen--qwen3-vl-30b-a3b-instruct","qwen/qwen3-vl-30b-a3b-instruct (on Novita AI)","throughput_tps","tokens per second",115.399,17,102.208,120.622,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-122b-a10b","qwen/qwen3.5-122b-a10b (on Novita AI)","latency_ttft_ms","ms",824.4,17,660.8,1011.4,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-122b-a10b","qwen/qwen3.5-122b-a10b (on Novita AI)","throughput_tps","tokens per second",90.901,17,77.323,123.241,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-27b","qwen/qwen3.5-27b (on Novita AI)","latency_ttft_ms","ms",1028.2,17,822.0,2581.0,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-27b","qwen/qwen3.5-27b (on Novita AI)","throughput_tps","tokens per second",51.842,17,44.475,69.043,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-35b-a3b","qwen/qwen3.5-35b-a3b (on Novita AI)","latency_ttft_ms","ms",1025.0,17,710.6,1476.0,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-35b-a3b","qwen/qwen3.5-35b-a3b (on Novita AI)","throughput_tps","tokens per second",75.495,17,44.037,88.903,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-397b-a17b","qwen/qwen3.5-397b-a17b (on Novita AI)","latency_ttft_ms","ms",1035.4,17,830.4,1661.8,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.5-397b-a17b","qwen/qwen3.5-397b-a17b (on Novita AI)","throughput_tps","tokens per second",71.416,17,68.452,78.248,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.8-2.4t-a95b","qwen/qwen3.8-2.4t-a95b (on Novita AI)","latency_ttft_ms","ms",1391.0,17,1025.6,1969.6,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.8-2.4t-a95b","qwen/qwen3.8-2.4t-a95b (on Novita AI)","throughput_tps","tokens per second",32.604,17,27.449,39.257,"2026-09-02","2026-09-18"],["novita","qwen--qwen3.8-27b","qwen/qwen3.8-27b (on Novita AI)","latency_ttft_ms","ms",633.4,10,469.4,730.0,"2026-09-09","2026-09-18"],["novita","qwen--qwen3.8-27b","qwen/qwen3.8-27b (on Novita AI)","throughput_tps","tokens per second",38.536,10,28.568,41.804,"2026-09-09","2026-09-18"],["novita","sao10k--l3-8b-stheno-v3.2","Sao10K/L3-8B-Stheno-v3.2 (on Novita AI)","latency_ttft_ms","ms",502.4,17,386.2,882.6,"2026-09-02","2026-09-18"],["novita","sao10k--l3-8b-stheno-v3.2","Sao10K/L3-8B-Stheno-v3.2 (on Novita AI)","throughput_tps","tokens per second",98.59,17,88.823,107.181,"2026-09-02","2026-09-18"],["novita","thudm--glm-4-32b-0414","thudm/glm-4-32b-0414 (on Novita AI)","latency_ttft_ms","ms",828.225,16,612.2,1004.4,"2026-09-03","2026-09-18"],["novita","thudm--glm-4-32b-0414","thudm/glm-4-32b-0414 (on Novita AI)","throughput_tps","tokens per second",50.763,16,37.67,57.156,"2026-09-03","2026-09-18"],["novita","xiaomimimo--mimo-v2.5","xiaomimimo/mimo-v2.5 (on Novita AI)","latency_ttft_ms","ms",1496.6,17,804.2,4452.25,"2026-09-02","2026-09-18"],["novita","xiaomimimo--mimo-v2.5","xiaomimimo/mimo-v2.5 (on Novita AI)","throughput_tps","tokens per second",31.248,17,12.874,48.954,"2026-09-02","2026-09-18"],["novita","xiaomimimo--mimo-v2.5-pro","xiaomimimo/mimo-v2.5-pro (on Novita AI)","latency_ttft_ms","ms",2293.0,17,1038.6,4923.0,"2026-09-02","2026-09-18"],["novita","xiaomimimo--mimo-v2.5-pro","xiaomimimo/mimo-v2.5-pro (on Novita AI)","throughput_tps","tokens per second",28.487,17,21.45,39.695,"2026-09-02","2026-09-18"],["novita","zai-org--autoglm-phone-9b-multilingual","zai-org/autoglm-phone-9b-multilingual (on Novita AI)","latency_ttft_ms","ms",522.2,17,397.4,910.2,"2026-09-02","2026-09-18"],["novita","zai-org--autoglm-phone-9b-multilingual","zai-org/autoglm-phone-9b-multilingual (on Novita AI)","throughput_tps","tokens per second",92.187,17,77.722,116.122,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.5-air","zai-org/glm-4.5-air (on Novita AI)","latency_ttft_ms","ms",645.2,17,494.2,887.2,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.5-air","zai-org/glm-4.5-air (on Novita AI)","throughput_tps","tokens per second",76.214,17,39.584,98.158,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.5v","zai-org/glm-4.5v (on Novita AI)","latency_ttft_ms","ms",1059.4,17,648.0,1941.2,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.5v","zai-org/glm-4.5v (on Novita AI)","throughput_tps","tokens per second",39.397,17,14.302,79.903,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.6","zai-org/glm-4.6 (on Novita AI)","latency_ttft_ms","ms",1314.0,17,759.6,1854.6,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.6","zai-org/glm-4.6 (on Novita AI)","throughput_tps","tokens per second",45.6,17,38.211,61.214,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.7","zai-org/glm-4.7 (on Novita AI)","latency_ttft_ms","ms",1677.0,17,1104.2,2137.8,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.7","zai-org/glm-4.7 (on Novita AI)","throughput_tps","tokens per second",42.639,17,35.443,58.547,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.7-flash","zai-org/glm-4.7-flash (on Novita AI)","latency_ttft_ms","ms",1030.333,15,581.0,2318.5,"2026-09-02","2026-09-18"],["novita","zai-org--glm-4.7-flash","zai-org/glm-4.7-flash (on Novita AI)","throughput_tps","tokens per second",71.473,15,49.948,94.691,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5","zai-org/glm-5 (on Novita AI)","latency_ttft_ms","ms",1069.8,17,942.2,1234.2,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5","zai-org/glm-5 (on Novita AI)","throughput_tps","tokens per second",49.182,17,43.184,57.186,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5.2","zai-org/glm-5.2 (on Novita AI)","latency_ttft_ms","ms",1346.1,16,1010.6,4704.0,"2026-09-03","2026-09-18"],["novita","zai-org--glm-5.2","zai-org/glm-5.2 (on Novita AI)","throughput_tps","tokens per second",52.662,16,29.6,61.664,"2026-09-03","2026-09-18"],["novita","zai-org--glm-5.3","zai-org/glm-5.3 (on Novita AI)","latency_ttft_ms","ms",1312.4,17,820.0,2245.2,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5.3","zai-org/glm-5.3 (on Novita AI)","throughput_tps","tokens per second",52.233,17,34.901,72.443,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5.3-flash","zai-org/glm-5.3-flash (on Novita AI)","latency_ttft_ms","ms",1257.667,17,920.0,1618.0,"2026-09-02","2026-09-18"],["novita","zai-org--glm-5.3-flash","zai-org/glm-5.3-flash (on Novita AI)","throughput_tps","tokens per second",66.017,17,25.234,75.638,"2026-09-02","2026-09-18"],["ovhcloud","gpt-oss-120b","gpt-oss-120b (on OVHcloud)","latency_ttft_ms","ms",312.8,17,242.2,380.6,"2026-09-02","2026-09-18"],["ovhcloud","gpt-oss-120b","gpt-oss-120b (on OVHcloud)","throughput_tps","tokens per second",95.493,17,64.104,137.639,"2026-09-02","2026-09-18"],["ovhcloud","gpt-oss-20b","gpt-oss-20b (on OVHcloud)","latency_ttft_ms","ms",314.8,17,245.4,512.6,"2026-09-02","2026-09-18"],["ovhcloud","gpt-oss-20b","gpt-oss-20b (on OVHcloud)","throughput_tps","tokens per second",65.628,17,40.727,81.614,"2026-09-02","2026-09-18"],["ovhcloud","qwen-2-5-vl-72b-instruct","Qwen2.5-VL-72B-Instruct (on OVHcloud)","latency_ttft_ms","ms",421.6,17,295.6,925.4,"2026-09-02","2026-09-18"],["ovhcloud","qwen-2-5-vl-72b-instruct","Qwen2.5-VL-72B-Instruct (on OVHcloud)","throughput_tps","tokens per second",32.453,17,23.34,35.128,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-5-397b","Qwen3.5-397B-A17B (on OVHcloud)","latency_ttft_ms","ms",411.2,18,295.8,1538.6,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-5-397b","Qwen3.5-397B-A17B (on OVHcloud)","throughput_tps","tokens per second",146.034,18,103.365,184.91,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-5-9b","Qwen3.5-9B (on OVHcloud)","latency_ttft_ms","ms",420.2,17,376.4,2748.4,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-5-9b","Qwen3.5-9B (on OVHcloud)","throughput_tps","tokens per second",38.233,17,27.143,39.124,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-6-27b","Qwen3.6-27B (on OVHcloud)","latency_ttft_ms","ms",372.0,17,302.4,1052.0,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-6-27b","Qwen3.6-27B (on OVHcloud)","throughput_tps","tokens per second",72.944,17,64.302,77.289,"2026-09-02","2026-09-18"],["ovhcloud","qwen-3-8-27b","Qwen3.8-27B (on OVHcloud)","latency_ttft_ms","ms",434.6,1,434.6,434.6,"2026-09-18","2026-09-18"],["ovhcloud","qwen-3-8-27b","Qwen3.8-27B (on OVHcloud)","throughput_tps","tokens per second",61.02,1,61.02,61.02,"2026-09-18","2026-09-18"],["zai","glm-4-5","GLM-4.5","latency_ttft_ms","ms",1518.4,10,1352.6,2463.75,"2026-09-02","2026-09-16"],["zai","glm-4-5","GLM-4.5","throughput_tps","tokens per second",44.213,10,38.85,61.971,"2026-09-02","2026-09-16"],["zai","glm-4-5-air","GLM-4.5-Air","latency_ttft_ms","ms",1439.2,17,448.2,2036.2,"2026-09-02","2026-09-18"],["zai","glm-4-5-air","GLM-4.5-Air","throughput_tps","tokens per second",74.263,17,35.852,101.15,"2026-09-02","2026-09-18"],["zai","glm-4-5v","GLM-4.5V","latency_ttft_ms","ms",2739.8,17,661.4,3857.8,"2026-09-02","2026-09-18"],["zai","glm-4-5v","GLM-4.5V","throughput_tps","tokens per second",39.774,17,23.27,73.654,"2026-09-02","2026-09-18"],["zai","glm-4-6","GLM-4.6","latency_ttft_ms","ms",1006.0,11,792.8,2596.2,"2026-09-02","2026-09-18"],["zai","glm-4-6","GLM-4.6","throughput_tps","tokens per second",38.296,11,20.177,47.656,"2026-09-02","2026-09-18"],["zai","glm-4-6v","GLM-4.6V","latency_ttft_ms","ms",2075.4,17,535.4,3284.0,"2026-09-02","2026-09-18"],["zai","glm-4-6v","GLM-4.6V","throughput_tps","tokens per second",52.408,17,28.368,70.884,"2026-09-02","2026-09-18"],["zai","glm-4-6v-flash","GLM-4.6V-Flash","latency_ttft_ms","ms",2515.5,15,597.0,4670.5,"2026-09-02","2026-09-18"],["zai","glm-4-6v-flash","GLM-4.6V-Flash","throughput_tps","tokens per second",69.585,15,33.568,90.967,"2026-09-02","2026-09-18"],["zai","glm-4-7","GLM-4.7","latency_ttft_ms","ms",1357.708,8,990.6,3079.5,"2026-09-02","2026-09-11"],["zai","glm-4-7","GLM-4.7","throughput_tps","tokens per second",37.499,8,29.878,49.4,"2026-09-02","2026-09-11"],["zai","glm-4-7-flash","GLM-4.7-Flash","latency_ttft_ms","ms",1296.0,17,494.0,3363.5,"2026-09-02","2026-09-18"],["zai","glm-4-7-flash","GLM-4.7-Flash","throughput_tps","tokens per second",81.116,17,53.953,123.426,"2026-09-02","2026-09-18"],["zai","glm-5","GLM-5","latency_ttft_ms","ms",1021.1,10,851.667,3881.0,"2026-09-02","2026-09-18"],["zai","glm-5","GLM-5","throughput_tps","tokens per second",56.988,10,52.049,63.773,"2026-09-02","2026-09-18"],["zai","glm-5-1","GLM-5.1","latency_ttft_ms","ms",3924.667,9,3147.333,4164.5,"2026-09-02","2026-09-15"],["zai","glm-5-1","GLM-5.1","throughput_tps","tokens per second",26.947,9,24.513,27.827,"2026-09-02","2026-09-15"],["zai","glm-5-2","GLM-5.2","latency_ttft_ms","ms",2205.0,17,895.2,2753.0,"2026-09-02","2026-09-18"],["zai","glm-5-2","GLM-5.2","throughput_tps","tokens per second",53.972,17,49.36,61.35,"2026-09-02","2026-09-18"],["zai","glm-5-3","GLM-5.3","latency_ttft_ms","ms",2242.1,18,980.8,3233.8,"2026-09-02","2026-09-18"],["zai","glm-5-3","GLM-5.3","throughput_tps","tokens per second",51.272,18,32.517,58.555,"2026-09-02","2026-09-18"],["zai","glm-5-3-flash","GLM-5.3-Flash","latency_ttft_ms","ms",1636.6,17,932.2,2425.4,"2026-09-02","2026-09-18"],["zai","glm-5-3-flash","GLM-5.3-Flash","throughput_tps","tokens per second",64.09,17,30.855,75.861,"2026-09-02","2026-09-18"]]}