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 |
| 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. | $0.25 | — |
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. | $1.50 | — |
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. | $0.025 | — |
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. | $0.125Google AI Studio | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.75Google AI Studio | — |
| $125Google 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. | 72% | 83.9% |
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. | 16% | 24% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 86.9% | 87.8% |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 73.2% | 78.4% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 76.8% | 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. | 11% | — |
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. | 38.3% | — |
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% |
MLE-BenchML engineering — Can the AI do the work of a machine-learning engineer? It competes in real Kaggle competitions — building, training, and tuning models end to end — and the score reflects how well it places. Higher is better. | 22% | — |
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 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | 31% | — |
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% |
AIME 2026Reasoning — single-answer accuracy on the 2026 invitational mathematics exam without tools | — | 94.1% |
SimpleQA VerifiedKnowledge — accuracy on the vetted factual questions in SimpleQA Verified | 43.3% | — |
MMMLUKnowledge — average accuracy on multilingual translations of MMLU | 88.9% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 86.2% |
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 54.3% | — |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | 642 | — |
CharXiv Reasoning · with toolsChart reasoning — The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better. | 75.6% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 82.9% |
MRCR v2 (8-needle) · 128k averageLong context — Tests whether the AI can find specific details buried inside a very long document (around 128k tokens — roughly a long book). Higher is better. | 60.1% | — |
MRCR v2 (8-needle) · 1M pointwiseLong context — Tests whether the AI can find specific details buried inside an enormous document (around 1 million tokens — many books). Higher is better. | 12.3% | — |
| Overview | ||
| Company | Qwen | |
| Release date | Mar 3 2026 | Apr 22 2026 |
| Access | Closed | Open Weight |
| Model details | View model | View model |
Other comparisons
Gemini 3.1 Flash-LitevsClaude Fable 5.2Qwen3.6-27BvsClaude Fable 5.2Gemini 3.1 Flash-LitevsGPT-6.2 AstraQwen3.6-27BvsGPT-6.2 AstraGemini 3.1 Flash-LitevsMuse EmberQwen3.6-27BvsMuse EmberGemini 3.1 Flash-LitevsGrok 4.7Qwen3.6-27BvsGrok 4.7Gemini 3.1 Flash-LitevsDeepSeek-V4.1-Flash-1017Qwen3.6-27BvsDeepSeek-V4.1-Flash-1017Gemini 3.1 Flash-LitevsMistral Large 4Qwen3.6-27BvsMistral Large 4Frequently asked questions
Qwen3.6-27B leads Gemini 3.1 Flash-Lite on 4 of the 5 benchmarks they both report. Only Gemini 3.1 Flash-Lite has a verified first-party price for an available API: $0.25 per million input tokens and $1.50 per million output tokens. No currently available pay-as-you-go API rate is tracked for Qwen3.6-27B. Gemini 3.1 Flash-Lite shipped 50 days before Qwen3.6-27B, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (Gemini 3.1 Flash-Lite) vs 256k (Qwen3.6-27B). Gemini 3.1 Flash-Lite is closed, while Qwen3.6-27B is open weight.
On LiveCodeBench, Qwen3.6-27B leads at 83.9% vs Gemini 3.1 Flash-Lite at 72%. On Humanity's Last Exam · no tools, Qwen3.6-27B leads at 24% vs Gemini 3.1 Flash-Lite at 16%. On GPQA Diamond, Qwen3.6-27B leads at 87.8% vs Gemini 3.1 Flash-Lite at 86.9%. On CharXiv Reasoning, Qwen3.6-27B leads at 78.4% vs Gemini 3.1 Flash-Lite at 73.2%. On MMMU-Pro, Gemini 3.1 Flash-Lite leads at 76.8% vs Qwen3.6-27B at 75.8%.