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. | 24B | >1T |
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 | 256k |
| 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. | 70.88% | 85.9% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 70.07% | 87.4% |
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. | — | 75.3% |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | — | 30.2% |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | — | 49.8% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | 86.14% | — |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 77.34% | — |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 70.9% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 85.7% |
| Overview | ||
| Company | Mistral | Qwen |
| Release date | Sep 17 2025 | Jan 26 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Magistral Small 1.2vsClaude Fable 5.2Qwen3-Max-ThinkingvsClaude Fable 5.2Magistral Small 1.2vsGPT-6.2 AstraQwen3-Max-ThinkingvsGPT-6.2 AstraMagistral Small 1.2vsGemini 4 NeonQwen3-Max-ThinkingvsGemini 4 NeonMagistral Small 1.2vsMuse EmberQwen3-Max-ThinkingvsMuse EmberMagistral Small 1.2vsGrok 4.7Qwen3-Max-ThinkingvsGrok 4.7Magistral Small 1.2vsDeepSeek-V4.1-Flash-1017Qwen3-Max-ThinkingvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3-Max-Thinking leads Magistral Small 1.2 on 2 of the 2 benchmarks they both report (LiveCodeBench, GPQA Diamond). Magistral Small 1.2 shipped 131 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.
Magistral Small 1.2 has 24B parameters, while Qwen3-Max-Thinking has >1T. Context windows are 128k (Magistral Small 1.2) vs 256k (Qwen3-Max-Thinking). Magistral Small 1.2 is open weight, while Qwen3-Max-Thinking is closed.
On LiveCodeBench, Qwen3-Max-Thinking leads at 85.9% vs Magistral Small 1.2 at 70.88%. On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs Magistral Small 1.2 at 70.07%.