Filters:
| Select | Model ↕ | Creator ↕ | Context Window ↕ | Intelligence ↓ⓘ | Price ($/M) ↕ⓘ | Speed (Tokens/s) ↕ | Latency (First Chunk s) ↕ | End-to-End Response (s) ↕ |
|---|---|---|---|---|---|---|---|---|
Claude Opus 5 (max) claude-opus-5-max | 1M | 61.00 | $2.03/M | 54 tok/s | 50.61 s | 59.92 s | ||
Claude Opus 5 (xhigh) claude-opus-5-xhigh | 1M | 60.00 | $1.56/M | 51 tok/s | 39.35 s | 49.16 s | ||
Claude Fable 5 (with fallback) claude-fable-5-with-fallback | 1M | 60.00 | $2.75/M | 73 tok/s | 123.99 s | 130.83 s | ||
GPT-5.6 Sol (max) gpt-5-6-sol-max | 1M | 59.00 | $1.54/M | 71 tok/s | 143.85 s | 150.94 s | ||
Claude Opus 5 (high) claude-opus-5-high | 1M | 59.00 | $1.06/M | 50 tok/s | 13.61 s | 23.62 s | ||
GPT-5.6 Sol (xhigh) gpt-5-6-sol-xhigh | 1M | 58.00 | $0.94/M | 69 tok/s | 33.92 s | 41.22 s | ||
Kimi K3 kimi-k3 | Kimi | 1.1M | 57.00 | $0.72/M | 32 tok/s | 4.44 s | 81.51 s | |
Claude Opus 5 (medium) claude-opus-5-medium | 1M | 56.00 | $0.62/M | 47 tok/s | 5.78 s | 16.35 s | ||
GPT-5.6 Sol (high) gpt-5-6-sol-high | 1M | 56.00 | $0.62/M | 68 tok/s | 11.58 s | 18.95 s | ||
GPT-5.6 Terra (max) gpt-5-6-terra-max | 1M | 55.00 | $0.78/M | 139 tok/s | 163.13 s | 166.73 s | ||
Grok 4.5 (high) grok-4-5-high | SpaceXAI | 500k | 54.00 | $0.35/M | 52 tok/s | 8.10 s | 17.71 s | |
GPT-5.6 Sol (medium) gpt-5-6-sol-medium | 1M | 54.00 | $0.41/M | 71 tok/s | 4.13 s | 11.19 s | ||
Claude Sonnet 5 (max) claude-sonnet-5-max | 1M | 53.00 | $1.53/M | 75 tok/s | 148.26 s | 154.94 s | ||
GPT-5.6 Terra (xhigh) gpt-5-6-terra-xhigh | 1M | 52.00 | $0.45/M | 125 tok/s | 12.60 s | 16.61 s | ||
GPT-5.6 Luna (max) gpt-5-6-luna-max | 1M | 51.00 | $0.29/M | 197 tok/s | 124.31 s | 126.85 s | ||
GLM-5.2 (max) glm-5-2-max | Z AI | 1M | 51.00 | $0.27/M | 198 tok/s | 1.36 s | 13.98 s | |
Muse Spark 1.1 (xhigh) muse-spark-1-1-xhigh | 1.1M | 51.00 | $0.26/M | 129 tok/s | 1.67 s | 21.01 s | ||
Claude Opus 5 (low) claude-opus-5-low | 1M | 51.00 | $0.36/M | 48 tok/s | 3.03 s | 13.47 s | ||
Gemini 3.5 Flash gemini-3-5-flash | 1M | 50.00 | $0.59/M | 195 tok/s | 22.56 s | 25.13 s | ||
Gemini 3.6 Flash gemini-3-6-flash | 1M | 50.00 | $0.50/M | 236 tok/s | 14.79 s | 16.91 s | ||
GPT-5.6 Sol (low) gpt-5-6-sol-low | 1M | 49.00 | $0.24/M | 66 tok/s | 3.48 s | 11.10 s | ||
GPT-5.6 Luna (xhigh) gpt-5-6-luna-xhigh | 1M | 49.00 | $0.19/M | 189 tok/s | 37.54 s | 40.19 s | ||
GPT-5.6 Terra (high) gpt-5-6-terra-high | 1M | 49.00 | $0.32/M | 113 tok/s | 1.97 s | 6.41 s | ||
Gemini 3.1 Pro Preview gemini-3-1-pro-preview | 1M | 46.00 | $0.29/M | 121 tok/s | 33.70 s | 37.82 s | ||
GPT-5.6 Luna (high) gpt-5-6-luna-high | 1M | 46.00 | $0.13/M | 203 tok/s | 7.33 s | 9.80 s | ||
Qwen3.7 Max qwen3-7-max | 1M | 46.00 | $1.03/M | 202 tok/s | 2.56 s | 16.94 s | ||
GPT-5.6 Terra (medium) gpt-5-6-terra-medium | 1M | 46.00 | $0.17/M | 117 tok/s | 1.54 s | 5.82 s | ||
Gemini 3.5 Flash (medium) gemini-3-5-flash-medium | 1M | 45.00 | — | 187 tok/s | 17.96 s | 20.63 s | ||
MiniMax-M3 minimax-m3 | MiniMax | 1M | 44.00 | $0.12/M | 83 tok/s | 1.46 s | 31.57 s | |
DeepSeek V4 Pro (max) deepseek-v4-pro-max | 1M | 44.00 | $0.04/M | 73 tok/s | 1.53 s | 68.12 s | ||
GPT-5.3 Codex (xhigh) gpt-5-3-codex-xhigh | 400k | 44.00 | — | 123 tok/s | 67.98 s | 72.03 s | ||
Motif 3 (Beta) motif-3-beta | Motif Technologies | 262k | 44.00 | — | — | — | — | |
DeepSeek V4 Pro (high) deepseek-v4-pro-high | 1M | 43.00 | $0.04/M | 72 tok/s | 1.57 s | 36.18 s | ||
Muse Spark muse-spark | 262k | 43.00 | — | — | — | — | ||
MiMo-V2.5-Pro mimo-v2-5-pro | Xiaomi | 1M | 42.00 | $0.04/M | 64 tok/s | 3.20 s | 42.34 s | |
Kimi K2.7 Code kimi-k2-7-code | Kimi | 256k | 42.00 | $0.20/M | 39 tok/s | 2.93 s | 73.40 s | |
Claude Sonnet 5 (Non-reasoning) claude-sonnet-5-non-reasoning | 1M | 42.00 | $0.37/M | 62 tok/s | 1.22 s | 9.34 s | ||
Hy3 hy3 | Tencent | 256k | 41.00 | $0.03/M | 62 tok/s | 2.64 s | 43.14 s | |
GPT-5.6 Sol (Non-reasoning) gpt-5-6-sol-non-reasoning | 1M | 41.00 | $0.26/M | 67 tok/s | 0.97 s | 8.45 s | ||
Nex-N2-Pro nex-n2-pro | Nex AGI | 262k | 41.00 | — | 131 tok/s | 1.77 s | 20.86 s | |
Inkling inkling | Thinking Machines | 1M | 41.00 | — | 83 tok/s | 1.76 s | 31.74 s | |
GPT-5.6 Terra (low) gpt-5-6-terra-low | 1M | 40.00 | $0.15/M | 124 tok/s | 1.29 s | 5.32 s | ||
DeepSeek V4 Flash (max) deepseek-v4-flash-max | 1M | 40.00 | $0.02/M | 117 tok/s | 1.26 s | 53.58 s | ||
Qwen3.6 Plus qwen3-6-plus | 1M | 40.00 | $0.31/M | 53 tok/s | 2.51 s | 116.45 s | ||
Qwen3.7 Plus qwen3-7-plus | 1M | 39.00 | $0.21/M | 53 tok/s | 2.85 s | 50.06 s | ||
JT-4.1 Flash 236B A21B jt-4-1-flash-236b-a21b | China Mobile | 256k | 39.00 | — | — | — | — | |
Agnes 2.5 Pro Alpha agnes-2-5-pro-alpha | Sapiens AI | 1M | 39.00 | — | 95 tok/s | 1.66 s | 27.88 s | |
GPT-5.6 Luna (medium) gpt-5-6-luna-medium | 1M | 38.00 | $0.07/M | 194 tok/s | 1.99 s | 4.57 s | ||
Nemotron 3 Ultra nemotron-3-ultra | NVIDIA | 262k | 38.00 | $0.38/M | 187 tok/s | 1.16 s | 16.03 s | |
DeepSeek V4 Flash (high) deepseek-v4-flash-high | 1M | 37.00 | $0.05/M | — | — | — | ||
MiMo-V2.5 mimo-v2-5 | Xiaomi | 1M | 37.00 | $0.01/M | 75 tok/s | 3.63 s | 36.75 s | |
Qwen3.6 27B qwen3-6-27b | 262k | 37.00 | $0.27/M | 56 tok/s | 3.66 s | 114.75 s | ||
Gemini 3.5 Flash-Lite gemini-3-5-flash-lite | 1M | 36.00 | $0.09/M | 409 tok/s | 7.79 s | 9.01 s | ||
MiMo-V2-Omni-0327 mimo-v2-omni-0327 | Xiaomi | 256k | 36.00 | — | — | — | — | |
Grok 4.3 (medium) grok-4-3-medium | SpaceXAI | 1M | 36.00 | — | 112 tok/s | 10.86 s | 15.31 s | |
Grok 4.3 (low) grok-4-3-low | SpaceXAI | 1M | 35.00 | — | 106 tok/s | 5.65 s | 10.36 s | |
MiMo-V2-Omni mimo-v2-omni | Xiaomi | 256k | 35.00 | — | — | — | — | |
Gemini 3.5 Flash (minimal) gemini-3-5-flash-minimal | 1M | 35.00 | — | 149 tok/s | 0.94 s | 4.29 s | ||
Kimi K2.6 kimi-k2-6 | Kimi | 256k | 35.00 | — | 37 tok/s | 2.90 s | 16.43 s | |
Claude Sonnet 4.6 (Non-reasoning, Low Effort) claude-sonnet-4-6-non-reasoning-low-effort | 1M | 34.00 | — | 42 tok/s | 1.26 s | 13.04 s | ||
GLM-5.2 glm-5-2 | Z AI | 1M | 34.00 | — | 87 tok/s | 1.55 s | 7.30 s | |
GPT-5.6 Terra (Non-reasoning) gpt-5-6-terra-non-reasoning | 1M | 34.00 | $0.17/M | 125 tok/s | 0.72 s | 4.72 s | ||
KAT-Coder-Pro V2 kat-coder-pro-v2 | KwaiKAT | 256k | 34.00 | — | 105 tok/s | 1.47 s | 6.23 s | |
Qwen3.5 397B A17B qwen3-5-397b-a17b | 262k | 34.00 | $0.33/M | 73 tok/s | 2.31 s | 52.69 s | ||
Hy3-preview hy3-preview | Tencent | 256k | 34.00 | — | 155 tok/s | 3.27 s | 19.40 s | |
LongCat 2.0 longcat-2-0 | LongCat | 1M | 33.00 | $0.11/M | 41 tok/s | 2.72 s | 63.50 s | |
GPT-5.6 Luna (low) gpt-5-6-luna-low | 1M | 33.00 | $0.06/M | 179 tok/s | 1.54 s | 4.33 s | ||
MiMo-V2-Flash (Feb 2026) mimo-v2-flash-feb-2026 | Xiaomi | 256k | 33.00 | — | — | — | — | |
Qwen3.5 122B A10B qwen3-5-122b-a10b | 262k | 32.00 | $0.24/M | 133 tok/s | 2.32 s | 21.09 s | ||
Qwen3.5 397B A17B qwen3-5-397b-a17b | 262k | 32.00 | — | 70 tok/s | 2.27 s | 9.43 s | ||
Qwen3.6 35B A3B qwen3-6-35b-a3b | 262k | 32.00 | $0.18/M | 151 tok/s | 2.24 s | 41.29 s | ||
DeepSeek V4 Pro deepseek-v4-pro | 1M | 31.00 | — | 71 tok/s | 1.57 s | 8.59 s | ||
Qwen3.5 Omni Plus qwen3-5-omni-plus | 256k | 31.00 | — | 53 tok/s | 2.41 s | 11.87 s | ||
Ring-2.6-1T ring-2-6-1t | InclusionAI | 262k | 31.00 | $0.35/M | 120 tok/s | 3.21 s | 23.97 s | |
Qwen3.6 27B qwen3-6-27b | 262k | 30.00 | $0.36/M | 56 tok/s | 3.65 s | 12.64 s | ||
o3 o3 | 200k | 30.00 | — | 128 tok/s | 5.73 s | 9.64 s | ||
Step 3.7 Flash step-3-7-flash | StepFun | 262k | 30.00 | $0.09/M | 400 tok/s | 0.92 s | 7.17 s | |
Mistral Medium 3.5 mistral-medium-3-5 | 256k | 30.00 | $0.56/M | 75 tok/s | 1.95 s | 35.33 s | ||
Claude 4.5 Haiku claude-4-5-haiku | 200k | 30.00 | $0.24/M | 100 tok/s | 14.57 s | 19.57 s | ||
Gemma 4 31B gemma-4-31b | 256k | 29.00 | $0.00/M | 35 tok/s | 1.12 s | 65.11 s | ||
GPT-5.5 Instant (June 2026) gpt-5-5-instant-june-2026 | 400k | 29.00 | $0.54/M | — | — | — | ||
DeepSeek V4 Flash deepseek-v4-flash | 1M | 29.00 | — | 113 tok/s | 1.19 s | 5.62 s | ||
JT-35B-Flash jt-35b-flash | China Mobile | 256k | 28.00 | — | — | — | — | |
KAT-Coder-Pro V1 kat-coder-pro-v1 | KwaiKAT | 256k | 28.00 | — | — | — | — | |
MiMo-V2.5-Pro mimo-v2-5-pro | Xiaomi | 1M | 28.00 | — | 62 tok/s | 2.58 s | 10.62 s | |
Qwen3.5 122B A10B qwen3-5-122b-a10b | 262k | 28.00 | $0.18/M | 150 tok/s | 2.35 s | 5.68 s | ||
GPT-5.6 Luna (Non-reasoning) gpt-5-6-luna-non-reasoning | 1M | 27.00 | $0.08/M | 180 tok/s | 0.73 s | 3.51 s | ||
Hy3-preview hy3-preview | Tencent | 256k | 26.00 | — | 164 tok/s | 3.16 s | 6.21 s | |
Ling-2.6-1T ling-2-6-1t | InclusionAI | 262k | 26.00 | — | — | — | — | |
Doubao Seed Code doubao-seed-code | ByteDance Seed | 256k | 26.00 | — | — | — | — | |
Gemini 2.5 Pro gemini-2-5-pro | 1M | 26.00 | $0.20/M | 144 tok/s | 22.27 s | 25.75 s | ||
Gemma 4 26B A4B gemma-4-26b-a4b | 256k | 26.00 | $0.04/M | — | — | — | ||
NVIDIA Nemotron 3 Super nvidia-nemotron-3-super | NVIDIA | 1M | 25.00 | $0.20/M | 153 tok/s | 1.70 s | 18.08 s | |
Gemini 3.1 Flash-Lite gemini-3-1-flash-lite | 1M | 25.00 | $0.04/M | 331 tok/s | 5.85 s | 7.36 s | ||
Grok 4.3 (Non-reasoning) grok-4-3-non-reasoning | SpaceXAI | 1M | 25.00 | $0.29/M | 110 tok/s | 0.75 s | 5.30 s | |
MiMo-V2-Flash mimo-v2-flash | Xiaomi | 256k | 25.00 | — | — | — | — | |
Qwen3.6 35B A3B qwen3-6-35b-a3b | 262k | 24.00 | $0.60/M | 173 tok/s | 2.26 s | 5.15 s | ||
Qwen3.5 35B A3B qwen3-5-35b-a3b | 262k | 24.00 | $0.23/M | 157 tok/s | 2.29 s | 5.48 s | ||
gpt-oss-120b (high) gpt-oss-120b-high | 131k | 24.00 | $0.06/M | 261 tok/s | 0.82 s | 10.40 s | ||
Claude 4.5 Haiku claude-4-5-haiku | 200k | 24.00 | — | 91 tok/s | 1.06 s | 6.53 s | ||
Command A+ command-a | 192k | 23.00 | $0.00/M | 200 tok/s | 0.41 s | 12.94 s | ||
K-EXAONE k-exaone | LG AI Research | 256k | 22.00 | — | — | — | — | |
ERNIE 5.0 Thinking Preview ernie-5-0-thinking-preview | Baidu | 128k | 22.00 | — | — | — | — | |
Gemma 4 12B gemma-4-12b | 256k | 22.00 | $0.07/M | 45 tok/s | 2.56 s | 58.68 s | ||
Gemma 4 31B gemma-4-31b | 256k | 22.00 | $0.03/M | 73 tok/s | 2.57 s | 9.42 s | ||
Nova 2.0 Pro Preview (medium) nova-2-0-pro-preview-medium | Amazon | 256k | 22.00 | $0.17/M | 116 tok/s | 13.34 s | 34.91 s | |
Qwen3.5 9B qwen3-5-9b | 262k | 21.00 | $0.22/M | 83 tok/s | 1.60 s | 31.89 s | ||
Mercury 2 mercury-2 | Inception | 128k | 21.00 | $0.08/M | 978 tok/s | 3.34 s | 3.85 s | |
Qwen3 Coder Next qwen3-coder-next | 256k | 21.00 | $0.33/M | 134 tok/s | 1.37 s | 5.10 s | ||
Nova 2.0 Omni (medium) nova-2-0-omni-medium | Amazon | 1M | 21.00 | — | — | — | — | |
Apriel-v1.6-15B-Thinker apriel-v1-6-15b-thinker | ServiceNow | 128k | 21.00 | — | — | — | — | |
Qwen3.5 9B qwen3-5-9b | 262k | 20.00 | — | — | — | — | ||
EXAONE 4.5 33B exaone-4-5-33b | LG AI Research | 262k | 20.00 | — | — | — | — | |
Gemma 4 26B A4B gemma-4-26b-a4b | 256k | 20.00 | — | 52 tok/s | 1.26 s | 10.79 s | ||
Qwen3.5 4B qwen3-5-4b | 262k | 20.00 | — | 19 tok/s | 0.76 s | 132.97 s | ||
North Mini Code north-mini-code | 256k | 20.00 | $0.00/M | 37 tok/s | 0.51 s | 68.13 s | ||
Nova 2.0 Pro Preview (low) nova-2-0-pro-preview-low | Amazon | 256k | 20.00 | $0.21/M | 116 tok/s | 9.19 s | 30.68 s | |
Mistral Small 4 mistral-small-4 | 256k | 20.00 | $0.10/M | 145 tok/s | 0.75 s | 17.94 s | ||
Devstral 2 devstral-2 | 256k | 19.00 | $0.00/M | 48 tok/s | 1.21 s | 11.55 s | ||
Nova 2.0 Lite (medium) nova-2-0-lite-medium | Amazon | 1M | 19.00 | — | 138 tok/s | 16.37 s | 34.54 s | |
Qwen3.5 Omni Flash qwen3-5-omni-flash | 256k | 19.00 | — | 236 tok/s | 1.86 s | 3.97 s | ||
JT-MINI jt-mini | China Mobile | 128k | 19.00 | — | — | — | — | |
Nova 2.0 Lite (high) nova-2-0-lite-high | Amazon | 1M | 18.00 | $0.25/M | 128 tok/s | 20.95 s | 40.55 s | |
Trinity Large Thinking trinity-large-thinking | Arcee AI | 512k | 18.00 | $0.15/M | 174 tok/s | 0.95 s | 15.34 s | |
Magistral Medium 1.2 magistral-medium-1-2 | 128k | 18.00 | $0.75/M | 45 tok/s | 1.74 s | 56.80 s | ||
Nova 2.0 Lite (low) nova-2-0-lite-low | Amazon | 1M | 18.00 | — | 152 tok/s | 9.28 s | 25.76 s | |
HyperNova 60B 2605 hypernova-60b-2605 | Multiverse Computing | 131k | 18.00 | $0.02/M | 356 tok/s | 0.89 s | 7.92 s | |
Nemotron Cascade 2 30B A3B nemotron-cascade-2-30b-a3b | NVIDIA | 1M | 18.00 | — | — | — | — | |
Devstral Small 2 devstral-small-2 | 256k | 17.00 | $0.00/M | 54 tok/s | 1.25 s | 10.53 s | ||
K2 Think V2 k2-think-v2 | MBZUAI Institute of Foundation Models | 262k | 17.00 | — | — | — | — | |
LongCat Flash Lite longcat-flash-lite | LongCat | 256k | 17.00 | — | — | — | — | |
HyperCLOVA X SEED Think (32B) hyperclova-x-seed-think-32b | Naver | 128k | 17.00 | — | — | — | — | |
K-EXAONE k-exaone | LG AI Research | 256k | 17.00 | — | — | — | — | |
Qwen3 Next 80B A3B qwen3-next-80b-a3b | 262k | 17.00 | $0.17/M | 207 tok/s | 2.25 s | 14.33 s | ||
Nova 2.0 Omni (low) nova-2-0-omni-low | Amazon | 1M | 17.00 | — | — | — | — | |
Mi:dm K 2.5 Pro mi-dm-k-2-5-pro | Korea Telecom | 128k | 16.00 | — | — | — | — | |
G9v3-3B g9v3-3b | AI9Stars | 131k | 16.00 | $0.00/M | — | — | — | |
Qwen3.5 4B qwen3-5-4b | 262k | 16.00 | — | 24 tok/s | 0.76 s | 21.65 s | ||
Mistral Large 3 mistral-large-3 | 256k | 16.00 | $0.06/M | 46 tok/s | 1.11 s | 11.90 s | ||
INTELLECT-3 intellect-3 | Prime Intellect | 131k | 16.00 | — | — | — | — | |
Solar Open 100B solar-open-100b | Upstage | 128k | 15.00 | — | — | — | — | |
Nemotron 3 Nano Omni 30B A3B Reasoning nemotron-3-nano-omni-30b-a3b-reasoning | NVIDIA | 256k | 15.00 | — | 319 tok/s | 0.96 s | 8.80 s | |
gpt-oss-120b (low) gpt-oss-120b-low | 131k | 15.00 | $0.02/M | 286 tok/s | 0.86 s | 9.58 s | ||
gpt-oss-20b (high) gpt-oss-20b-high | 131k | 15.00 | $0.02/M | 189 tok/s | 0.81 s | 14.06 s | ||
Nova 2.0 Pro Preview nova-2-0-pro-preview | Amazon | 256k | 14.00 | $0.25/M | 92 tok/s | 1.04 s | 6.48 s | |
gpt-oss-20b (low) gpt-oss-20b-low | 131k | 14.00 | — | 196 tok/s | 0.86 s | 13.63 s | ||
Llama 4 Maverick llama-4-maverick | 1M | 14.00 | $0.03/M | 104 tok/s | 0.94 s | 5.74 s | ||
K2-V2 (high) k2-v2-high | MBZUAI Institute of Foundation Models | 512k | 14.00 | — | — | — | — | |
NVIDIA Nemotron 3 Nano nvidia-nemotron-3-nano | NVIDIA | 1M | 14.00 | $0.02/M | 201 tok/s | 1.18 s | 13.59 s | |
Solar Pro 3 solar-pro-3 | Upstage | 128k | 14.00 | $0.11/M | 107 tok/s | 2.29 s | 25.57 s | |
Ling 2.6 Flash ling-2-6-flash | InclusionAI | 262k | 14.00 | — | 116 tok/s | 1.12 s | 5.45 s | |
Qwen3 Next 80B A3B qwen3-next-80b-a3b | 262k | 14.00 | — | 181 tok/s | 2.17 s | 4.93 s | ||
DiffusionGemma 26B A4B diffusiongemma-26b-a4b | 256k | 13.00 | — | — | — | — | ||
Gemma 4 12B (Non-reasoning) gemma-4-12b-non-reasoning | 262k | 13.00 | — | 34 tok/s | 2.53 s | 17.08 s | ||
Motif-2-12.7B motif-2-12-7b | Motif Technologies | 128k | 13.00 | — | — | — | — | |
Nova Premier nova-premier | Amazon | 1M | 13.00 | — | 33 tok/s | 2.87 s | 18.13 s | |
K2-V2 (medium) k2-v2-medium | MBZUAI Institute of Foundation Models | 512k | 12.00 | — | — | — | — | |
Llama Nemotron Super 49B v1.5 llama-nemotron-super-49b-v1-5 | NVIDIA | 128k | 12.00 | — | 77 tok/s | 5.95 s | 38.27 s | |
Mistral Small 4 mistral-small-4 | 256k | 12.00 | — | 147 tok/s | 0.70 s | 4.10 s | ||
Tri-21B-Think tri-21b-think | Trillion Labs | 32k | 12.00 | — | — | — | — | |
MiniCPM5-1B minicpm5-1b | OpenBMB | 128k | 12.00 | — | — | — | — | |
Sarvam 105B (high) sarvam-105b-high | Sarvam | 128k | 12.00 | — | — | — | — | |
Gemma 4 E4B gemma-4-e4b | 128k | 12.00 | — | 90 tok/s | 0.79 s | 28.52 s | ||
Nova 2.0 Lite nova-2-0-lite | Amazon | 1M | 12.00 | — | 139 tok/s | 1.12 s | 4.72 s | |
MiniCPM5-1B minicpm5-1b | OpenBMB | 128k | 12.00 | — | — | — | — | |
Magistral Small 1.2 magistral-small-1-2 | 128k | 11.00 | $0.25/M | 88 tok/s | 0.94 s | 29.32 s | ||
Nanbeige4.1-3B nanbeige4-1-3b | Nanbeige | 256k | 11.00 | — | — | — | — | |
Ministral 3 14B ministral-3-14b | 256k | 11.00 | $0.15/M | 76 tok/s | 0.84 s | 7.44 s | ||
EXAONE 4.0 32B exaone-4-0-32b | LG AI Research | 131k | 11.00 | — | — | — | — | |
Nova 2.0 Omni nova-2-0-omni | Amazon | 1M | 11.00 | — | — | — | — | |
Llama 4 Scout llama-4-scout | 10M | 10.00 | $0.01/M | 84 tok/s | 0.78 s | 6.73 s | ||
Hermes 4 70B hermes-4-70b | Nous Research | 128k | 10.00 | — | 90 tok/s | 1.35 s | 29.14 s | |
Falcon-H1R-7B falcon-h1r-7b | TII UAE | 256k | 10.00 | — | — | — | — | |
Qwen3 Omni 30B A3B qwen3-omni-30b-a3b | 66k | 10.00 | — | 99 tok/s | 1.94 s | 27.15 s | ||
Step3 VL 10B step3-vl-10b | StepFun | 66k | 9.00 | — | — | — | — | |
Gemma 4 E2B gemma-4-e2b | 128k | 9.00 | — | — | — | — | ||
Llama 3.3 70B llama-3-3-70b | 128k | 9.00 | $0.08/M | 83 tok/s | 1.65 s | 7.70 s | ||
Llama Nemotron Ultra llama-nemotron-ultra | NVIDIA | 128k | 9.00 | — | 53 tok/s | 2.31 s | 49.84 s | |
ERNIE 4.5 300B A47B ernie-4-5-300b-a47b | Baidu | 131k | 9.00 | — | — | — | — | |
Hermes 4 405B hermes-4-405b | Nous Research | 128k | 9.00 | — | 37 tok/s | 2.40 s | 70.51 s | |
NVIDIA Nemotron Nano 12B v2 VL nvidia-nemotron-nano-12b-v2-vl | NVIDIA | 128k | 9.00 | — | 79 tok/s | 8.49 s | 40.06 s | |
Ministral 3 8B ministral-3-8b | 256k | 9.00 | $0.18/M | 118 tok/s | 0.73 s | 4.97 s | ||
Gemma 4 E4B gemma-4-e4b | 128k | 9.00 | — | 92 tok/s | 0.76 s | 6.19 s | ||
Granite 4.1 30B granite-4-1-30b | IBM | 131k | 9.00 | — | — | — | — | |
NVIDIA Nemotron Nano 9B V2 nvidia-nemotron-nano-9b-v2 | NVIDIA | 131k | 9.00 | — | 107 tok/s | 4.58 s | 27.94 s | |
Hermes 4 405B hermes-4-405b | Nous Research | 128k | 9.00 | — | 37 tok/s | 2.38 s | 15.85 s | |
NVIDIA Nemotron 3 Nano 4B nvidia-nemotron-3-nano-4b | NVIDIA | 262k | 9.00 | — | — | — | — | |
Llama Nemotron Super 49B v1.5 llama-nemotron-super-49b-v1-5 | NVIDIA | 128k | 9.00 | — | 67 tok/s | 4.54 s | 11.96 s | |
K2-V2 (low) k2-v2-low | MBZUAI Institute of Foundation Models | 512k | 9.00 | — | — | — | — | |
Kimi Linear 48B A3B Instruct kimi-linear-48b-a3b-instruct | Kimi | 1M | 9.00 | — | — | — | — | |
Llama 3.1 405B llama-3-1-405b | 128k | 9.00 | — | — | — | — | ||
LFM2.5-8B-A1B lfm2-5-8b-a1b | Liquid AI | 33k | 8.00 | — | 339 tok/s | 1.96 s | 9.34 s | |
Ring-flash-2.0 ring-flash-2-0 | InclusionAI | 128k | 8.00 | — | — | — | — | |
Olmo 3.1 32B Think olmo-3-1-32b-think | Allen Institute for AI | 66k | 8.00 | — | — | — | — | |
Command A command-a | 256k | 8.00 | — | 60 tok/s | 1.88 s | 10.17 s | ||
Llama 3.1 Nemotron 70B llama-3-1-nemotron-70b | NVIDIA | 128k | 8.00 | — | 73 tok/s | 6.46 s | 13.27 s | |
NVIDIA Nemotron 3 Nano nvidia-nemotron-3-nano | NVIDIA | 1M | 7.00 | — | 115 tok/s | 0.91 s | 5.25 s | |
NVIDIA Nemotron Nano 9B V2 nvidia-nemotron-nano-9b-v2 | NVIDIA | 131k | 7.00 | — | 162 tok/s | 1.60 s | 4.69 s | |
Qwen3.5 2B qwen3-5-2b | 262k | 7.00 | — | — | — | — | ||
Hermes 4 70B hermes-4-70b | Nous Research | 128k | 7.00 | — | 95 tok/s | 1.34 s | 6.62 s | |
Granite 4.1 8B granite-4-1-8b | IBM | 131k | 7.00 | — | 98 tok/s | 0.79 s | 5.87 s | |
Sarvam 30B (high) sarvam-30b-high | Sarvam | 66k | 7.00 | — | — | — | — | |
Olmo 3.1 32B Instruct olmo-3-1-32b-instruct | Allen Institute for AI | 66k | 6.00 | — | — | — | — | |
Gemma 4 E2B gemma-4-e2b | 128k | 6.00 | — | — | — | — | ||
R1 1776 r1-1776 | 128k | 6.00 | — | — | — | — | ||
Ministral 3 3B ministral-3-3b | 256k | 6.00 | $0.13/M | 253 tok/s | 0.61 s | 2.59 s | ||
Llama 3.2 90B (Vision) llama-3-2-90b-vision | 128k | 6.00 | — | — | — | — | ||
Phi-4 Mini phi-4-mini | Microsoft | 128k | 6.00 | $0.00/M | 45 tok/s | 0.83 s | 11.89 s | |
EXAONE 4.0 32B exaone-4-0-32b | LG AI Research | 131k | 6.00 | — | — | — | — | |
Qwen3.5 2B qwen3-5-2b | 262k | 6.00 | — | — | — | — | ||
Qwen3.5 0.8B qwen3-5-0-8b | 262k | 5.00 | — | — | — | — | ||
DeepHermes 3 - Mistral 24B deephermes-3-mistral-24b | Nous Research | 32k | 5.00 | — | — | — | — | |
Jamba 1.7 Large jamba-1-7-large | AI21 Labs | 256k | 5.00 | — | 55 tok/s | 1.42 s | 10.54 s | |
Granite 4.0 H Small granite-4-0-h-small | IBM | 128k | 5.00 | — | 365 tok/s | 10.23 s | 11.60 s | |
Qwen3 Omni 30B A3B qwen3-omni-30b-a3b | 66k | 5.00 | — | 95 tok/s | 1.89 s | 7.17 s | ||
LFM2 24B A2B lfm2-24b-a2b | Liquid AI | 33k | 5.00 | — | — | — | — | |
Phi-4 phi-4 | Microsoft | 16k | 5.00 | — | 29 tok/s | 4.01 s | 21.17 s | |
Nova Micro nova-micro | Amazon | 130k | 5.00 | — | 268 tok/s | 0.96 s | 2.83 s | |
Granite 4.1 3B granite-4-1-3b | IBM | 131k | 5.00 | — | — | — | — | |
NVIDIA Nemotron Nano 12B v2 VL nvidia-nemotron-nano-12b-v2-vl | NVIDIA | 128k | 5.00 | — | 165 tok/s | 2.43 s | 5.47 s | |
Phi-4 Multimodal phi-4-multimodal | Microsoft | 128k | 5.00 | — | 18 tok/s | 0.81 s | 29.15 s | |
MiniCPM-V 4.6 1.3B minicpm-v-4-6-1-3b | OpenBMB | 262k | 4.00 | — | — | — | — | |
Jamba Reasoning 3B jamba-reasoning-3b | AI21 Labs | 262k | 4.00 | — | — | — | — | |
Reka Flash 3 reka-flash-3 | Reka AI | 128k | 4.00 | — | — | — | — | |
Olmo 3 7B Think olmo-3-7b-think | Allen Institute for AI | 66k | 4.00 | — | — | — | — | |
Molmo 7B-D molmo-7b-d | Allen Institute for AI | 4k | 4.00 | — | — | — | — | |
Ling-mini-2.0 ling-mini-2-0 | InclusionAI | 131k | 4.00 | — | — | — | — | |
Llama 3.2 11B (Vision) llama-3-2-11b-vision | 128k | 3.00 | — | 5 tok/s | 5.64 s | 103.98 s | ||
Qwen3.5 0.8B qwen3-5-0-8b | 262k | 3.00 | — | — | — | — | ||
Exaone 4.0 1.2B exaone-4-0-1-2b | LG AI Research | 64k | 3.00 | — | — | — | — | |
Olmo 3 7B olmo-3-7b | Allen Institute for AI | 66k | 3.00 | — | — | — | — | |
Exaone 4.0 1.2B exaone-4-0-1-2b | LG AI Research | 64k | 3.00 | — | — | — | — | |
LFM2.5-1.2B-Thinking lfm2-5-1-2b-thinking | Liquid AI | 32k | 3.00 | — | — | — | — | |
Jamba 1.7 Mini jamba-1-7-mini | AI21 Labs | 258k | 3.00 | — | — | — | — | |
LFM2 2.6B lfm2-2-6b | Liquid AI | 33k | 3.00 | — | — | — | — | |
LFM2.5-1.2B-Instruct lfm2-5-1-2b-instruct | Liquid AI | 32k | 3.00 | — | — | — | — | |
Granite 4.0 H 1B granite-4-0-h-1b | IBM | 128k | 3.00 | — | — | — | — | |
Gemma 3 270M gemma-3-270m | 32k | 2.00 | — | — | — | — | ||
Apertus 70B Instruct apertus-70b-instruct | Swiss AI Initiative | 66k | 2.00 | — | — | — | — | |
Granite 4.0 Micro granite-4-0-micro | IBM | 128k | 2.00 | — | — | — | — | |
DeepHermes 3 - Llama-3.1 8B deephermes-3-llama-3-1-8b | Nous Research | 128k | 2.00 | — | — | — | — | |
Granite 4.0 1B granite-4-0-1b | IBM | 128k | 2.00 | — | — | — | — | |
Molmo2-8B molmo2-8b | Allen Institute for AI | 37k | 2.00 | — | — | — | — | |
LFM2 8B A1B lfm2-8b-a1b | Liquid AI | 33k | 2.00 | — | — | — | — | |
LFM2.5-VL-1.6B lfm2-5-vl-1-6b | Liquid AI | 32k | 1.00 | — | 410 tok/s | 1.62 s | 2.84 s | |
Granite 4.0 350M granite-4-0-350m | IBM | 33k | 1.00 | — | — | — | — | |
Tiny Aya Global tiny-aya-global | 8k | 1.00 | — | — | — | — | ||
Apertus 8B Instruct apertus-8b-instruct | Swiss AI Initiative | 66k | 1.00 | — | — | — | — | |
Granite 4.0 H 350M granite-4-0-h-350m | IBM | 33k | 1.00 | — | — | — | — | |
Claude Sonnet 5 (low) claude-sonnet-5-low | 1M | — | — | 58 tok/s | 1.94 s | 10.63 s | ||
EXAONE 4.5 33B exaone-4-5-33b | LG AI Research | 262k | — | — | — | — | — | |
Claude Sonnet 5 (xhigh) claude-sonnet-5-xhigh | 1M | — | — | 69 tok/s | 24.31 s | 31.57 s | ||
Gemini 3 Deep Think gemini-3-deep-think | 128k | — | — | — | — | — | ||
Claude Sonnet 5 (high) claude-sonnet-5-high | 1M | — | — | 59 tok/s | 6.99 s | 15.44 s | ||
Mi:dm K 2.5 Pro Preview mi-dm-k-2-5-pro-preview | Korea Telecom | 128k | — | — | — | — | — | |
GPT-5.5 Pro (xhigh) gpt-5-5-pro-xhigh | 922k | — | — | — | — | — | ||
Cogito v2.1 cogito-v2-1 | Deep Cogito | 128k | — | — | — | — | — | |
Claude Sonnet 5 (medium) claude-sonnet-5-medium | 1M | — | — | 59 tok/s | 2.32 s | 10.86 s |