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. | 671B | 7B |
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 | 32k |
| 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. | $0.25DeepInfra | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.95DeepInfra | — |
| $97.50DeepInfra · fp4 | — | |
| These models have no shared benchmark scores. | ||
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. | 66% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | 54.5% | — |
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. | 74.8% | — |
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. | — | 38.4% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 30% | — |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | 15.9% | — |
MATHReasoning — accuracy on the original competition mathematics test set | — | 13.2% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 93.1% | — |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 88.4% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 80.1% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 84.8% | — |
| Overview | ||
| Company | DeepSeek | Mistral |
| Release date | Aug 21 2025 | May 22 2024 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek V3.1vsClaude Fable 5.2Mistral 7B v0.3vsClaude Fable 5.2DeepSeek V3.1vsGPT-6.2 AstraMistral 7B v0.3vsGPT-6.2 AstraDeepSeek V3.1vsGemini 4 HydrogenMistral 7B v0.3vsGemini 4 HydrogenDeepSeek V3.1vsMuse EmberMistral 7B v0.3vsMuse EmberDeepSeek V3.1vsGrok 4.7Mistral 7B v0.3vsGrok 4.7DeepSeek V3.1vsKimi K3.1 ProMistral 7B v0.3vsKimi K3.1 ProFrequently asked questions
DeepSeek V3.1 and Mistral 7B v0.3 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Mistral 7B v0.3 shipped 456 days before DeepSeek V3.1, so benchmark comparisons should account for the intervening progress.
DeepSeek V3.1 has 671B parameters, while Mistral 7B v0.3 has 7B. Context windows are 128k (DeepSeek V3.1) vs 32k (Mistral 7B v0.3).
Direct benchmark comparisons are unavailable — DeepSeek V3.1 and Mistral 7B v0.3 don't publish scores on any of the same benchmarks.