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. | — | >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. | 1M | 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. | 37.8% | 85.9% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 64.5% | 85.7% |
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% |
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. | 86.6% | — |
IFEvalAgentic tasks — following verifiable instructions in generated responses | 76.3% | — |
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% |
MATHReasoning — accuracy on the original competition mathematics test set | 81.1% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 87.4% |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 70.9% |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | 93.8% | — |
| Overview | ||
| Company | Qwen | Qwen |
| Release date | Nov 15 2024 | Jan 26 2026 |
| Access | Closed | Closed |
| Model details | View model | View model |
Other comparisons
Qwen2.5-Turbo (1M)vsClaude Fable 5.2Qwen3-Max-ThinkingvsClaude Fable 5.2Qwen2.5-Turbo (1M)vsGPT-6.2 AstraQwen3-Max-ThinkingvsGPT-6.2 AstraQwen2.5-Turbo (1M)vsGemini 4 NeonQwen3-Max-ThinkingvsGemini 4 NeonQwen2.5-Turbo (1M)vsMuse EmberQwen3-Max-ThinkingvsMuse EmberQwen2.5-Turbo (1M)vsGrok 4.7Qwen3-Max-ThinkingvsGrok 4.7Qwen2.5-Turbo (1M)vsDeepSeek-V4.1-Flash-1017Qwen3-Max-ThinkingvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3-Max-Thinking leads Qwen2.5-Turbo (1M) on 2 of the 2 benchmarks they both report (LiveCodeBench, MMLU-Pro). Qwen2.5-Turbo (1M) shipped 437 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Qwen2.5-Turbo (1M)) vs 256k (Qwen3-Max-Thinking).
On LiveCodeBench, Qwen3-Max-Thinking leads at 85.9% vs Qwen2.5-Turbo (1M) at 37.8%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs Qwen2.5-Turbo (1M) at 64.5%.