Model Browser
Compare large language models side by side. Filter GPT-4o, Claude, Gemini, Llama, Mistral, DeepSeek, and Qwen models by intelligence benchmarks like GPQA Diamond, HLE, and LiveCodeBench. Sort by API pricing, tokens per second, and overall score. Find the best LLM for your use case.
0
0
0
0 tok/s
50
$50 /1M
Intelligenceby
All Models
| Model | Intelligence | Cost | Speed |
|---|---|---|---|
| Gemini 3.1 Pro | ~81 | $4.50 | 130tok/s |
| GPT-5.4 | ~78 | $5.63 | 87tok/s |
| Gemini 3 Pro | 76 | $4.50 | 142tok/s |
| Gemini 3 Flash | 75 | $1.13 | 215tok/s |
| GPT-5.2 | 74 | $4.81 | 88tok/s |
| Claude Opus 4.6 | 72 | $10.00 | 73tok/s |
| GPT-5.1 | 69 | $3.44 | 136tok/s |
| Claude Opus 4.5 | 69 | $10.00 | 88tok/s |
| Kimi K2.5 | 69 | $1.07 | 345tok/s |
| Qwen3.5 397B | 68 | $1.35 | 74tok/s |
| GLM-4.7 | 68 | $1.00 | 136tok/s |
| GPT-5 | 67 | $3.44 | 98tok/s |
| Grok 4 | 66 | $6.00 | 36tok/s |
| Claude Sonnet 4.6 | 65 | $6.00 | 64tok/s |
| DeepSeek V3.2 | 65 | $0.32 | 219tok/s |
| GPT-5 mini | 63 | $0.69 | 75tok/s |
| o3 | 62 | $3.50 | 141tok/s |
| Gemini 2.5 Pro | 62 | $3.44 | 159tok/s |
| Grok 4.1 Fast | 62 | $0.28 | 150tok/s |
| GPT-OSS 120B | 61 | $0.26 | 2951tok/s |
| o4-mini | 60 | $1.93 | 134tok/s |
| Claude Sonnet 4.5 | 57 | $6.00 | 103tok/s |
| DeepSeek R1 | 57 | $2.36 | 306tok/s |
| MiniMax M2.5 | 57 | $0.52 | 273tok/s |
| DeepSeek V3.1 | 55 | $0.84 | 347tok/s |
| GLM-5 | 55 | $1.55 | 266tok/s |
| Gemini 2.5 Flash | 52 | $0.85 | 282tok/s |
| Magistral Medium | 51 | $2.75 | 29tok/s |
| GPT-5 nano | 49 | $0.14 | 131tok/s |
| o1 | 48 | $26.25 | 174tok/s |
| Claude Haiku 4.5 | 37 | $2.00 | 109tok/s |
| Mistral Large 3 | 37 | $0.75 | 147tok/s |
| GPT-4.1 | 36 | $3.50 | 104tok/s |
| GPT-4.1 mini | 36 | $0.70 | 78tok/s |
| Llama 4 Maverick | 34 | $0.30 | 434tok/s |
| DeepSeek V3 | 28 | $0.52 | 120tok/s |
| GPT-4.1 nano | 24 | $0.18 | 142tok/s |
| Command A | 24 | $4.38 | 49tok/s |
| GPT-4o | 23 | $4.38 | 170tok/s |
| GPT-4o mini | 17 | $0.26 | 54tok/s |
| Command R+ | 9 | $4.38 | 59tok/s |