Speed & Latency
Output speed (tokens per second) and latency (time to first token) benchmarks across models and providers.
Speed vs Intelligence
EUGlobal
Axes stay fixed while filtering · Drag to zoom · Click to pin68 selected of 68 chartable models. Axes stay fixed for comparison.
85 models
| # | |||||||
|---|---|---|---|---|---|---|---|
| 1 | Gemini 3.1 Flash-LiteGlobal | - | 341.9 t/s | 4.94s | 25 | $0.56 | |
| 2 | GPT-OSS 120BOpenEU | OpenAI | Nebius | 324.6 t/s | 0.51s | 23.8 | $0.26 |
| 3 | Gemini 2.5 Flash-LiteEU | Google Cloud | 255.9 t/s | 19.93s | 11.4 | $0.17 | |
| 4 | GPT-5.6 LunaRGlobal | OpenAI | - | 254.7 t/s | 71.64s | 51.2 | $2.25 |
| 5 | Gemini 3.5 FlashRGlobal | - | 241.5 t/s | 13.01s | 50.2 | $3.38 | |
| 6 | o3-miniREU | OpenAI | Azure | 241.1 t/s | 16.66s | 15.6 | $1.93 |
| 7 | Nemotron 3 SuperROpenGlobal | NVIDIA | - | 233.2 t/s | 1.04s | 25.4 | $0.38 |
| 8 | Gemini 2.5 FlashREU | Google Cloud | 228.4 t/s | 9.71s | 20.1 | $0.85 | |
| 9 | GPT-4oEU | OpenAI | Azure | 223.2 t/s | 0.52s | 11.2 | $4.38 |
| 10 | GLM-5.2REU | Z AI | Tensorix | 216.7 t/s | 0.84s | 51.1 | $2.25 |
| 11 | Gemini 3 FlashRGlobal | - | 216.5 t/s | 5.09s | 37.8 | $1.13 | |
| 12 | o4-miniREU | OpenAI | Azure | 194.7 t/s | 21.29s | 25.6 | $1.93 |
| 13 | Ministral 3 3BOpenGlobal | Mistral | - | 182.4 t/s | 0.38s | 6.8 | $0.10 |
| 14 | GPT-5.4 NanoRGlobal | OpenAI | - | 178.9 t/s | 2.80s | 38.2 | $0.46 |
| 15 | Grok 4.20RGlobal | xAI | - | 177.5 t/s | 19.88s | 37 | $1.56 |
| 16 | GPT-5.4 MiniRGlobal | OpenAI | - | 176.8 t/s | 5.77s | 40 | $1.69 |
| 17 | Nova 2 LiteREU | Amazon | Bedrock | 175.9 t/s | 14.47s | 18.2 | $0.85 |
| 18 | Mistral Small 4ROpenGlobal | Mistral | - | 175.9 t/s | 0.49s | 19.6 | $0.15 |
| 19 | GPT-5.6 TerraRGlobal | OpenAI | - | 172.1 t/s | 108.76s | 55 | $5.63 |
| 20 | GPT-5.4RGlobal | OpenAI | - | 171.1 t/s | 103.53s | 51.4 | $5.63 |
| 21 | GPT-5 NanoREU | OpenAI | Azure | 166.7 t/s | 88.10s | 19.9 | $0.14 |
| 22 | Grok 4.3RGlobal | xAI | - | 158.9 t/s | 36.58s | 37.6 | $1.56 |
| 23 | o3REU | OpenAI | Azure | 158.8 t/s | 5.01s | 30.4 | $3.50 |
| 24 | GPT-4.1 NanoEU | OpenAI | Azure | 154.5 t/s | 0.49s | 9.6 | $0.17 |
| 25 | Gemini 3.1 ProRGlobal | - | 147.9 t/s | 27.38s | 46.5 | $4.50 | |
| 26 | GPT-4.1EU | OpenAI | Azure | 145.5 t/s | 0.64s | 19.4 | $3.50 |
| 27 | Gemini 2.5 ProREU | Google Cloud | 145.2 t/s | 17.23s | 25.8 | $3.44 | |
| 28 | o1REU | OpenAI | Azure | 139 t/s | 18.74s | 23.4 | $26.25 |
| 29 | Claude Haiku 4.5REU | Anthropic | Bedrock | 137.2 t/s | 12.83s | 29.6 | $2.00 |
| 30 | GPT-5.1REU | OpenAI | Azure | 126.3 t/s | 33.56s | 36.9 | $3.44 |
| 31 | Llama 4 MaverickOpenGlobal | Meta | - | 121.7 t/s | 0.62s | 14.3 | $0.47 |
| 32 | Grok 4.5RGlobal | xAI | - | 121 t/s | 15.57s | 53.8 | $3.00 |
| 33 | MiniMax M3REU | MiniMax | Tensorix | 116.1 t/s | 1.47s | 44.4 | $0.80 |
| 34 | DeepSeek V4 FlashREU | DeepSeek | Tensorix | 114.3 t/s | 0.87s | 40.3 | $0.19 |
| 35 | GPT-5REU | OpenAI | Azure | 110.7 t/s | 87.39s | 34.7 | $3.44 |
| 36 | GPT-5.3 CodexRGlobal | OpenAI | - | 107.3 t/s | 53.86s | 44.3 | $4.81 |
| 37 | Ministral 3 8BOpenGlobal | Mistral | - | 106.1 t/s | 0.39s | 9 | $0.15 |
| 38 | GPT-5 MiniREU | OpenAI | Azure | 101.7 t/s | 74.06s | 25.3 | $0.69 |
| 39 | Hermes 4 70BROpenEU | NousResearch | Nebius | 96.5 t/s | 0.58s | 10 | $0.20 |
| 40 | Llama 3.3 70BOpenGlobal | Meta | - | 94.7 t/s | 0.65s | 9.4 | $0.61 |
| 41 | GPT-5.6 SolRGlobal | OpenAI | - | 90 t/s | 84.12s | 58.9 | $11.25 |
| 42 | GPT-5.2RGlobal | OpenAI | - | 88.3 t/s | 104.61s | 42.2 | $4.81 |
| 43 | Ministral 3 14BOpenGlobal | Mistral | - | 87.3 t/s | 0.43s | 11.1 | $0.20 |
| 44 | Claude Sonnet 5RGlobal | Anthropic | - | 83.2 t/s | 109.12s | 53.4 | $4.00 |
| 45 | MiniMax M2.1REU | MiniMax | Nebius | 81.9 t/s | 1.16s | 31.4 | $0.53 |
| 46 | GPT-5.5RGlobal | OpenAI | - | 81.1 t/s | 29.26s | 54.8 | $11.25 |
| 47 | GPT-4.1 MiniEU | OpenAI | Azure | 80.9 t/s | 0.60s | 14.8 | $0.70 |
| 48 | GLM-5.1RGlobal | Z AI | - | 79.7 t/s | 0.88s | 40.2 | $2.15 |
| 49 | MiniMax M2.5RGlobal | MiniMax | - | 79.2 t/s | 1.28s | 33.7 | $0.53 |
| 50 | Mistral Medium 3.5Global | Mistral | - | 75 t/s | 0.72s | 29.9 | $3.00 |
| 51 | GPT-4o MiniEU | OpenAI | Azure | 68 t/s | 0.59s | 6.9 | $0.26 |
| 52 | Claude Fable 5RGlobal | Anthropic | - | 68 t/s | 73.28s | 59.9 | $20.00 |
| 53 | DeepSeek V4 ProREU | DeepSeek | Tensorix | 67.7 t/s | 1.04s | 44.3 | $2.19 |
| 54 | Qwen3 235BROpenEU | Alibaba | Nebius | 67 t/s | 1.13s | 13.4 | $2.63 |
| 55 | Claude Sonnet 4.6REU | Anthropic | Bedrock | 66.3 t/s | 80.44s | 47.2 | $6.00 |
| 56 | Claude Opus 4.5REU | Anthropic | Bedrock | 63.7 t/s | 13.68s | 40.8 | $10.00 |
| 57 | Mistral Large 3OpenGlobal | Mistral | - | 63 t/s | 0.60s | 15.9 | $0.75 |
| 58 | Qwen3.5 397BROpenGlobal | Alibaba | - | 61.5 t/s | 1.82s | 33.7 | $1.35 |
| 59 | Claude Opus 4.8REU | Anthropic | Bedrock | 60.9 t/s | 25.41s | 55.7 | $10.00 |
| 60 | Claude Sonnet 4.5REU | Anthropic | Bedrock | 55.2 t/s | 11.57s | 36.4 | $6.00 |
| 61 | Claude Opus 4.7REU | Anthropic | Bedrock | 53.6 t/s | 22.26s | 53.5 | $10.00 |
| 62 | Qwen3.6 PlusRGlobal | Alibaba | - | 52.7 t/s | 1.51s | 39.6 | $1.13 |
| 63 | Claude Opus 4.6REU | Anthropic | Bedrock | 51.7 t/s | 18.49s | 43.7 | $10.00 |
| 64 | GLM-5RGlobal | Z AI | - | 51 t/s | 0.75s | 39.5 | $1.55 |
| 65 | Kimi K2.5RGlobal | Kimi | - | 49.6 t/s | 1.08s | 38.1 | $1.20 |
| 66 | MiniMax M2.7RGlobal | MiniMax | - | 48.3 t/s | 1.47s | 38.1 | $0.53 |
| 67 | Grok 3 MiniRGlobal | xAI | - | 47.2 t/s | 0.67s | 22.5 | $0.35 |
| 68 | Kimi K2.7 CodeREU | Kimi | Tensorix | 46.4 t/s | 1.17s | 41.9 | $2.06 |
| 69 | Hermes 4 405BROpenEU | NousResearch | Nebius | 41.7 t/s | 0.73s | 9 | $1.50 |
| 70 | Claude Sonnet 4REU | Anthropic | Bedrock | - | - | 28.9 | $6.00 |
| 71 | Claude 3.7 SonnetREU | Anthropic | Bedrock | - | - | 27.1 | $6.00 |
| 72 | Gemini 3 ProRGlobal | - | - | - | 39.6 | $4.50 | |
| 73 | Muse SparkRGlobal | Meta | - | - | - | 43.1 | $0.00 |
| 74 | Grok 4RGlobal | xAI | - | - | - | 33.3 | $11.00 |
| 75 | DeepSeek V3.2ROpenGlobal | DeepSeek | - | - | - | 32 | $0.32 |
| 76 | DeepSeek V3.2 (Non-reasoning)OpenGlobal | DeepSeek | - | - | - | 24.7 | $0.32 |
| 77 | MiMo V2 ProROpenGlobal | Xiaomi | - | - | - | 40.3 | $1.50 |
| 78 | SonarGlobal | Perplexity | - | - | - | 9.5 | $1.00 |
| 79 | Sonar ProGlobal | Perplexity | - | - | - | 9.3 | $6.00 |
| 80 | Sonar Reasoning ProRGlobal | Perplexity | - | - | - | 17.8 | $3.50 |
| 81 | Magistral MediumRGlobal | Mistral | - | - | - | 12.5 | $2.75 |
| 82 | Magistral SmallRGlobal | Mistral | - | - | - | 10.7 | $0.75 |
| 83 | GPT-5.4 ProRGlobal | OpenAI | - | - | - | - | $67.50 |
| 84 | GPT-5.5 ProRGlobal | OpenAI | - | - | - | - | $67.50 |
| 85 | Grok 3Global | xAI | - | - | - | 18.4 | $8.00 |