Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
Comparing agents instead? Switch to agent compare →
 Gemini 2.5 Pro |  Qwen3 235BAlibaba | Gemini 3.5 |  DeepSeek-V4-ProDeepSeek | |
|---|---|---|---|---|
| Vendor | Alibaba | DeepSeek | ||
| Family | Gemini | Qwen | Gemini | DeepSeek |
| Release date | 2025-03-25 | 2025-04-29 | 2026-05-20 | 2026-04-22 |
| Context window | 2,000,000 tokens | 128,000 tokens | 2,000,000 tokens | 1,000,000 tokens |
| Parameters | — | 235B | — | 1.6T (49B active) |
| Modality | text, vision, audio, video | text | text, vision, audio, video | text |
| License | proprietary | Apache-2.0 | proprietary | MIT |
| Source | proprietary | open weights | proprietary | open weights |
| Description | Predecessor to Gemini 3 Pro. | Predecessor to the Qwen3.6 family. | Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026. | 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. |
| Links | ||||
| Benchmarks | ||||
| MMLU-Pro | 84.1% | — | 88.9% | 84.2% |
| GPQA-D | 84.0% | — | 91.2% | 82.4% |
| HumanEval | — | — | — | 95.1% |
| Aider | — | — | — | 80.1% |
| AIME-25 | — | — | 96.1% | 88.6% |
| LiveCB | — | — | — | 72.4% |
| MMMU | 72.4% | — | 86.0% | — |
Add a model
Claude 4.8 OpusmaxQwen3.7-MaxmaxGemini 3.5✓Gemini Omnimax
GPT-5.5maxQwen3.6-27BmaxDeepSeek-V4-FlashmaxDeepSeek-V4-Pro✓Gemini 3 PromaxClaude 4.7 OpusmaxClaude 4.6 SonnetmaxClaude 4.5 HaikumaxDeepSeek-V3.1maxGPT-5max
Grok 4maxGemini 2.5 FlashmaxQwen3 235B✓o3maxGPT-4.1max
Llama 4 ScoutmaxLlama 4 MaverickmaxGemini 2.5 Pro✓DeepSeek-R1max
Mistral Large 2maxMax 4 models. Remove one to add another.