Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
Comparing agents instead? Switch to agent compare →
 Qwen3.6-27BAlibaba |  GPT-5 ProOpenAI | Qwen3 235B Alibaba |  DeepSeek-V3.1DeepSeek | |
|---|---|---|---|---|
| Vendor | Alibaba | OpenAI | Alibaba | DeepSeek |
| Family | Qwen | GPT | Qwen | DeepSeek |
| Release date | 2026-04-22 | 2025-08-07 | 2025-04-29 | 2025-08-21 |
| Context window | 262,144 tokens | 400,000 tokens | 128,000 tokens | 128,000 tokens |
| Parameters | 27B (dense) | — | 235B | 671B |
| Modality | text, vision, video | text, vision, audio | text | text |
| License | Apache-2.0 | proprietary | Apache-2.0 | MIT |
| Source | open weights | proprietary | open weights | open weights |
| Description | Alibaba's first dense open-weight in the 3.6 family. Strong agentic-coding scores (77.2 SWE-bench Verified, matching Claude 4.5 Opus on Terminal-Bench 2.0). Supports 201 languages and multimodal text/image/video input. | Higher-reasoning variant of GPT-5, used for harder scientific and research tasks (e.g., immunology breakthroughs at NIH). Same 400K context as GPT-5; trades latency for depth. | Predecessor to the Qwen3.6 family. | Large MoE open-weight model. Predecessor to DeepSeek-V4. |
| Links | ||||
| Benchmarks | ||||
| MMLU-Pro | 78.9% | — | — | — |
| HumanEval | 91.5% | — | — | — |
| Aider | 70.2% | — | — | — |
| MMMU | 70.3% | — | — | — |
Add a model
GPT-5.6 Solmax
Claude 4.8 OpusmaxQwen3.7-Maxmax
Gemini 3.5 FlashmaxGemini 3.5maxGemini OmnimaxGPT-5.5maxDeepSeek-V4-FlashmaxGemini 3 PromaxDeepSeek-V4-PromaxQwen3.6-27B✓Claude 4.7 OpusmaxClaude 4.6 SonnetmaxClaude 4.5 HaikumaxDeepSeek-V3.1✓GPT-5 Pro✓GPT-5max
Grok 4maxGemini 2.5 FlashmaxQwen3 235B✓o3maxGPT-4.1max
Llama 4 ScoutmaxLlama 4 MaverickmaxGemini 2.5 PromaxDeepSeek-R1max
Mistral Large 2maxMax 4 models. Remove one to add another.