Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
 DeepSeek-V4-ProDeepSeek | DeepSeek-V3.1 DeepSeek |  Claude 4.6 SonnetAnthropic |  Qwen3 235BAlibaba | |
|---|---|---|---|---|
| Vendor | DeepSeek | DeepSeek | Anthropic | Alibaba |
| Family | DeepSeek | DeepSeek | Claude | Qwen |
| Release date | 2026-04-22 | 2025-08-21 | — | 2025-04-29 |
| Context window | 1,000,000 tokens | 128,000 tokens | 200,000 tokens | 128,000 tokens |
| Parameters | 1.6T (49B active) | 671B | — | 235B |
| Modality | text | text | text, vision | text |
| License | MIT | MIT | proprietary | Apache-2.0 |
| Source | open weights | open weights | proprietary | open weights |
| Description | DeepSeek's flagship open-weight MoE. 1.6T parameters with 49B activated, 1M-token context, and a hybrid attention scheme (CSA + HCA) that delivers long-context inference at ~27% of V3.2's FLOPs. | Large MoE open-weight model. Predecessor to DeepSeek-V4. | — | Predecessor to the Qwen3.6 family. |
| Links |
Add a model
Gemini 2.5 FlashmaxClaude 4.6 Sonnet✓
GPT-5.5maxDeepSeek-V4-FlashmaxQwen3.6-27BmaxGemini 3 PromaxDeepSeek-V4-Pro✓Claude 4.7 OpusmaxClaude 4.5 HaikumaxDeepSeek-V3.1✓GPT-5max
Grok 4maxQwen3 235B✓o3maxGPT-4.1max
Llama 4 ScoutmaxLlama 4 MaverickmaxGemini 2.5 PromaxDeepSeek-R1max
Mistral Large 2maxMax 4 models. Remove one to add another.