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. | — | 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. | 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. | 30% | 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. | 51% | 87.8% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 67.3% | 86.2% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 62.3% | 82.9% |
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% |
MATHReasoning — accuracy on the original competition mathematics test set | 77.9% | — |
AIME 2026Reasoning — single-answer accuracy on the 2026 invitational mathematics exam without tools | — | 94.1% |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 78.4% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 75.8% |
| Overview | ||
| Company | Qwen | |
| Release date | Sep 24 2024 | Apr 22 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Gemini 1.5 Flash-002vsClaude Fable 5.2Qwen3.6-27BvsClaude Fable 5.2Gemini 1.5 Flash-002vsGPT-6.2 AstraQwen3.6-27BvsGPT-6.2 AstraGemini 1.5 Flash-002vsMuse EmberQwen3.6-27BvsMuse EmberGemini 1.5 Flash-002vsGrok 4.7Qwen3.6-27BvsGrok 4.7Gemini 1.5 Flash-002vsDeepSeek-V4.1-Flash-1017Qwen3.6-27BvsDeepSeek-V4.1-Flash-1017Gemini 1.5 Flash-002vsMistral Large 4Qwen3.6-27BvsMistral Large 4Frequently asked questions
Qwen3.6-27B leads Gemini 1.5 Flash-002 on 4 of the 4 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU-Pro, MMMU). Gemini 1.5 Flash-002 shipped 575 days before Qwen3.6-27B, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Gemini 1.5 Flash-002) vs 256k (Qwen3.6-27B). Gemini 1.5 Flash-002 is closed, while Qwen3.6-27B is open weight.
On LiveCodeBench, Qwen3.6-27B leads at 83.9% vs Gemini 1.5 Flash-002 at 30%. On GPQA Diamond, Qwen3.6-27B leads at 87.8% vs Gemini 1.5 Flash-002 at 51%. On MMLU-Pro, Qwen3.6-27B leads at 86.2% vs Gemini 1.5 Flash-002 at 67.3%. On MMMU, Qwen3.6-27B leads at 82.9% vs Gemini 1.5 Flash-002 at 62.3%.