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. | 35B | — |
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 | 1M |
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.375International (Singapore) | $0.03International (Singapore) |
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. | $2.25 | $0.13 |
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. | $0.05Darkbloom | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.70Darkbloom | — |
| $45.00Darkbloom · fp4 | $12.50Qwen list price | |
| These models have no shared benchmark scores. | ||
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. | 49.5% | — |
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. | 73.4% | — |
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. | 67.2% | — |
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. | 29.4% | — |
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. | 80.4% | — |
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. | 51.5% | — |
MCP AtlasMulti-step tool use — Can the AI chain together many tools and steps to complete one bigger task, rather than doing just a single thing? Higher is better. | 62.8% | — |
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. | 21.4% | — |
AIME 2026Reasoning — single-answer accuracy on the 2026 invitational mathematics exam without tools | 92.7% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 86% | — |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 85.2% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 78% | — |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 75.3% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 81.7% | — |
| Overview | ||
| Company | Qwen | Qwen |
| Release date | Apr 16 2026 | Jul 21 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Qwen3.6vsClaude Fable 5.2Qwen3.7-FlashvsClaude Fable 5.2Qwen3.6vsGPT-6.2 AstraQwen3.7-FlashvsGPT-6.2 AstraQwen3.6vsGemini 4 NeonQwen3.7-FlashvsGemini 4 NeonQwen3.6vsMuse EmberQwen3.7-FlashvsMuse EmberQwen3.6vsGrok 4.7Qwen3.7-FlashvsGrok 4.7Qwen3.6vsDeepSeek-V4.1-Flash-1017Qwen3.7-FlashvsDeepSeek-V4.1-Flash-1017Frequently asked questions
Qwen3.6 and Qwen3.7-Flash don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Qwen3.7-Flash is cheaper on both input and output: $0.03 vs $0.375 per million input tokens, and $0.13 vs $2.25 per million output tokens. Figures are base-tier rates. The listed Qwen3.6 rates use the “International (Singapore)” pricing scope. The listed Qwen3.7-Flash rates use the “International (Singapore)” pricing scope. Qwen3.6 shipped 96 days before Qwen3.7-Flash, so benchmark comparisons should account for the intervening progress.
Context windows are 256k (Qwen3.6) vs 1M (Qwen3.7-Flash). Qwen3.6 is open weight, while Qwen3.7-Flash is closed.
Direct benchmark comparisons are unavailable — Qwen3.6 and Qwen3.7-Flash don't publish scores on any of the same benchmarks.