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. | — | 0.5B/1.5B/3B/7B/14B/32B |
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. | 256k | 128k |
| 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. | — | $0.66Cloudflare |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | — | $1.00Cloudflare |
| — | $182Cloudflare | |
| 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. | 37.9% | 31.4% |
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. | 86.6% | 92.7% |
IFEvalAgentic tasks — following verifiable instructions in generated responses | — | 79.9% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 76.4% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 62.3% |
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | — | 77.6% |
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. | — | 93% |
| Overview | ||
| Company | Mistral | Qwen |
| Release date | Jan 13 2025 | Sep 19 2024 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Codestral 25.01vsClaude Fable 5.2Qwen2.5-CodervsClaude Fable 5.2Codestral 25.01vsGPT-6.2 AstraQwen2.5-CodervsGPT-6.2 AstraCodestral 25.01vsGemini 4 NeonQwen2.5-CodervsGemini 4 NeonCodestral 25.01vsMuse EmberQwen2.5-CodervsMuse EmberCodestral 25.01vsGrok 4.7Qwen2.5-CodervsGrok 4.7Codestral 25.01vsDeepSeek-V4.1-Flash-1017Qwen2.5-CodervsDeepSeek-V4.1-Flash-1017Frequently asked questions
Codestral 25.01 and Qwen2.5-Coder are evenly matched across the 2 benchmarks they both report (LiveCodeBench, HumanEval). Qwen2.5-Coder shipped 116 days before Codestral 25.01, so benchmark comparisons should account for the intervening progress.
Context windows are 256k (Codestral 25.01) vs 128k (Qwen2.5-Coder). Codestral 25.01 is closed, while Qwen2.5-Coder is open weight.
On LiveCodeBench, Codestral 25.01 leads at 37.9% vs Qwen2.5-Coder at 31.4%. On HumanEval, Qwen2.5-Coder leads at 92.7% vs Codestral 25.01 at 86.6%.