Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | — | 7B/14B |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 128k | 1M |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.44Mistral | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.20Mistral | — |
| $198Mistral | — | |
| Benchmarks | ||
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 30.3% | 38.6% |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | 92.1% | 88.4% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | 89.4% | 84.3% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 77.2% | 63.3% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 79.5% |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | 91% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 57.1% | — |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | — | 94.8% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 66.1% | — |
| Overview | ||
| Company | Mistral | Qwen |
| Release date | May 7 2025 | Jan 27 2025 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral Medium 3vsClaude Fable 5.2Qwen2.5-1MvsClaude Fable 5.2Mistral Medium 3vsGPT-6.2 AstraQwen2.5-1MvsGPT-6.2 AstraMistral Medium 3vsGemini 4 NeonQwen2.5-1MvsGemini 4 NeonMistral Medium 3vsMuse EmberQwen2.5-1MvsMuse EmberMistral Medium 3vsGrok 4.7Qwen2.5-1MvsGrok 4.7Mistral Medium 3vsDeepSeek-V4.1-Flash-1017Qwen2.5-1MvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Mistral Medium 3 leads Qwen2.5-1M on 3 of the 4 benchmarks they both report (LiveCodeBench, HumanEval, IFEval, MMLU-Pro). Qwen2.5-1M shipped 100 days before Mistral Medium 3, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (Mistral Medium 3) vs 1M (Qwen2.5-1M). Mistral Medium 3 is closed, while Qwen2.5-1M is open weight.
On LiveCodeBench, Qwen2.5-1M leads at 38.6% vs Mistral Medium 3 at 30.3%. On HumanEval, Mistral Medium 3 leads at 92.1% vs Qwen2.5-1M at 88.4%. On IFEval, Mistral Medium 3 leads at 89.4% vs Qwen2.5-1M at 84.3%. On MMLU-Pro, Mistral Medium 3 leads at 77.2% vs Qwen2.5-1M at 63.3%.