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 | 125B |
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 | 262k |
| Benchmarks | ||
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. | 55.7% | 81% |
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | — | 62.5% |
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. | 68% | — |
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | — | 58.7% |
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. | — | 48.1% |
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. | — | 91.9% |
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. | 22.5% | — |
JobBenchProfessional tool use — Tests the AI on professional workplace tasks that require using real work tools — the kind of multi-step jobs an office worker handles. Higher is better. | — | 55.7% |
CoWorkBenchLong-horizon office work — Tests long-running office tasks across fields including computer science, finance, law, medicine, and other productivity work. Higher is better. | — | 73.9% |
Toolathlon-VerifiedPersonal tool use — Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better. | — | 73.5% |
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. | — | 35.9% |
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. | — | 35.9% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 91.7% |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 81.3% |
OSWorld 2.0Agentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Version 2.0 is a harder, refreshed task set. Higher is better. | — | 19.4% |
Agent's Last Exam · pass@1Agentic computer use — A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better. | — | 24.3% |
Agent's Last Exam · scoreAgentic computer use — The graded score on the same desktop and operating-system tasks in Agent's Last Exam, giving partial credit for progress beyond the strict pass-or-fail result. Higher is better. | — | 51.2% |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 84.6% |
LVBenchVideo understanding — Can the AI follow a very long video — up to an hour — and answer questions that need details from far apart in it? Higher is better. | — | 76.6% |
| Overview | ||
| Company | Mistral | Qwen |
| Release date | Dec 9 2025 | Aug 26 2026 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Devstral Small 2 (24B)vsClaude Haiku 5.5Qwen3.8-Flash-NextvsClaude Haiku 5.5Devstral Small 2 (24B)vsGPT-6.1 SolQwen3.8-Flash-NextvsGPT-6.1 SolDevstral Small 2 (24B)vsGemini 4 ArgonQwen3.8-Flash-NextvsGemini 4 ArgonDevstral Small 2 (24B)vsMuse Spark OpenQwen3.8-Flash-NextvsMuse Spark OpenDevstral Small 2 (24B)vsGrok 4.7Qwen3.8-Flash-NextvsGrok 4.7Devstral Small 2 (24B)vsDeepSeek-V4.1-Flash-1017Qwen3.8-Flash-NextvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3.8-Flash-Next leads Devstral Small 2 (24B) on 1 of the 1 benchmark they both report (SWE-Bench Multilingual). Devstral Small 2 (24B) shipped 260 days before Qwen3.8-Flash-Next, so benchmark comparisons should account for the intervening progress.
Devstral Small 2 (24B) has 24B parameters, while Qwen3.8-Flash-Next has 125B. Context windows are 256k (Devstral Small 2 (24B)) vs 262k (Qwen3.8-Flash-Next).
On SWE-Bench Multilingual, Qwen3.8-Flash-Next leads at 81% vs Devstral Small 2 (24B) at 55.7%.