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. | 119B | 27B |
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 | 256k |
| API pricingUSD per 1M tokens · lower wins | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.15 | — |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $0.60 | — |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.015 | — |
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. | $0.15Mistral | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.60Mistral | — |
| $60.00Mistral list price | — | |
| 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. | 63.6% | 83.9% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 71.2% | 87.8% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 78% | 86.2% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 60% | 75.8% |
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 6% | — |
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. | — | 77.2% |
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. | — | 71.3% |
NL2Repo-BenchRepo-level code generation — Tests whether the AI can turn a natural-language requirement into working code across an entire repository, not just produce a single function or patch. Higher is better. | — | 36.2% |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | — | 59.3% |
Humanity's Last Exam · no toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | — | 24% |
AIME 2026Reasoning — single-answer accuracy on the 2026 invitational mathematics exam without tools | — | 94.1% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 83.8% | — |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | 48% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 78.4% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 82.9% |
| Overview | ||
| Company | Mistral | Qwen |
| Release date | Mar 16 2026 | Apr 22 2026 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Mistral Small 4vsClaude Fable 5.2Qwen3.6-27BvsClaude Fable 5.2Mistral Small 4vsGPT-6.2 AstraQwen3.6-27BvsGPT-6.2 AstraMistral Small 4vsGemini 4 NeonQwen3.6-27BvsGemini 4 NeonMistral Small 4vsMuse EmberQwen3.6-27BvsMuse EmberMistral Small 4vsGrok 4.7Qwen3.6-27BvsGrok 4.7Mistral Small 4vsDeepSeek-V4.1-Flash-1017Qwen3.6-27BvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3.6-27B leads Mistral Small 4 on 4 of the 4 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU-Pro, MMMU-Pro). Only Mistral Small 4 has a verified first-party price for an available API: $0.15 per million input tokens and $0.60 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3.6-27B. Mistral Small 4 shipped 37 days before Qwen3.6-27B, so benchmark comparisons should account for the intervening progress.
Mistral Small 4 has 119B parameters, while Qwen3.6-27B has 27B. Context windows are 256k (Mistral Small 4) vs 256k (Qwen3.6-27B).
On LiveCodeBench, Qwen3.6-27B leads at 83.9% vs Mistral Small 4 at 63.6%. On GPQA Diamond, Qwen3.6-27B leads at 87.8% vs Mistral Small 4 at 71.2%. On MMLU-Pro, Qwen3.6-27B leads at 86.2% vs Mistral Small 4 at 78%. On MMMU-Pro, Qwen3.6-27B leads at 75.8% vs Mistral Small 4 at 60%.