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. | 236B | — |
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 | 40k |
| 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. | 41.8% | 59.35% |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 16.8% | — |
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. | 89% | — |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | — | 72.03% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | — | 60.99% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 71.46% |
| Overview | ||
| Company | DeepSeek | Mistral |
| Release date | Sep 5 2024 | Jul 24 2025 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek V2.5vsClaude Fable 5.2Magistral Medium 1.1vsClaude Fable 5.2DeepSeek V2.5vsGPT-6.2 AstraMagistral Medium 1.1vsGPT-6.2 AstraDeepSeek V2.5vsGemini 4 HydrogenMagistral Medium 1.1vsGemini 4 HydrogenDeepSeek V2.5vsMuse EmberMagistral Medium 1.1vsMuse EmberDeepSeek V2.5vsGrok 4.7Magistral Medium 1.1vsGrok 4.7DeepSeek V2.5vsKimi K3.1 ProMagistral Medium 1.1vsKimi K3.1 ProFrequently asked questions
Magistral Medium 1.1 leads DeepSeek V2.5 on 1 of the 1 benchmark they both report (LiveCodeBench). DeepSeek V2.5 shipped 322 days before Magistral Medium 1.1, so benchmark comparisons should account for the intervening progress.
Context windows are 128k (DeepSeek V2.5) vs 40k (Magistral Medium 1.1). DeepSeek V2.5 is open weight, while Magistral Medium 1.1 is closed.
On LiveCodeBench, Magistral Medium 1.1 leads at 59.35% vs DeepSeek V2.5 at 41.8%.