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. | 671B | >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. | 128k | 256k |
| API pricingUSD per 1M tokens · lower wins | ||
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. | $0.27SiliconFlow | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $1.00SiliconFlow | — |
| $104SiliconFlow · fp8 | — | |
| Benchmarks | ||
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. | 68.4% | 75.3% |
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. | 74.9% | 85.9% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 80.7% | 87.4% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 85% | 85.7% |
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. | 57.8% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 38.5% | — |
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% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | 88.4% | — |
IFBenchInstruction following — Tests whether the AI can follow detailed instructions and satisfy multiple constraints at once. Higher is better. | — | 70.9% |
| Overview | ||
| Company | DeepSeek | Qwen |
| Release date | Sep 22 2025 | Jan 26 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek V3.1 TerminusvsClaude Fable 5.2Qwen3-Max-ThinkingvsClaude Fable 5.2DeepSeek V3.1 TerminusvsGPT-6.2 AstraQwen3-Max-ThinkingvsGPT-6.2 AstraDeepSeek V3.1 TerminusvsGemini 4 NeonQwen3-Max-ThinkingvsGemini 4 NeonDeepSeek V3.1 TerminusvsMuse EmberQwen3-Max-ThinkingvsMuse EmberDeepSeek V3.1 TerminusvsGrok 4.7Qwen3-Max-ThinkingvsGrok 4.7DeepSeek V3.1 TerminusvsMistral Large 4Qwen3-Max-ThinkingvsMistral Large 4Frequently asked questions
Qwen3-Max-Thinking leads DeepSeek V3.1 Terminus on 4 of the 4 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, GPQA Diamond, MMLU-Pro). DeepSeek V3.1 Terminus shipped 126 days before Qwen3-Max-Thinking, so benchmark comparisons should account for the intervening progress.
DeepSeek V3.1 Terminus has 671B parameters, while Qwen3-Max-Thinking has >1T. Context windows are 128k (DeepSeek V3.1 Terminus) vs 256k (Qwen3-Max-Thinking). DeepSeek V3.1 Terminus is open weight, while Qwen3-Max-Thinking is closed.
On SWE-Bench Verified, Qwen3-Max-Thinking leads at 75.3% vs DeepSeek V3.1 Terminus at 68.4%. On LiveCodeBench, Qwen3-Max-Thinking leads at 85.9% vs DeepSeek V3.1 Terminus at 74.9%. On GPQA Diamond, Qwen3-Max-Thinking leads at 87.4% vs DeepSeek V3.1 Terminus at 80.7%. On MMLU-Pro, Qwen3-Max-Thinking leads at 85.7% vs DeepSeek V3.1 Terminus at 85%.