Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
Comparing agents instead? Switch to agent compare →
 DeepSeek-R1DeepSeek |  Claude 4.7 OpusAnthropic |  Qwen3 235BAlibaba |  Gemini 2.5 Flash | |
|---|---|---|---|---|
| Vendor | DeepSeek | Anthropic | Alibaba | |
| Family | DeepSeek | Claude | Qwen | Gemini |
| Release date | 2025-01-20 | 2026-04-16 | 2025-04-29 | 2025-06-17 |
| Context window | 128,000 tokens | 1,000,000 tokens | 128,000 tokens | 1,000,000 tokens |
| Parameters | 671B | — | 235B | — |
| Modality | text | text, vision | text | text, vision, audio |
| License | MIT | proprietary | Apache-2.0 | proprietary |
| Source | open weights | proprietary | open weights | proprietary |
| Description | Reasoning-focused open-weight model. | Anthropic's previous Opus flagship — superseded by Opus 4.8 on May 28, 2026 (42-day cycle). Optimized for complex reasoning and coding. Improved software engineering, long-running coding tasks, and higher-resolution vision over Claude 4.6. | Predecessor to the Qwen3.6 family. | Google's cost-and-latency-optimized Gemini 2.5 variant. Generally available June 2025 — designed for high-throughput agentic workflows with sub-second first-token latency. |
| Links | ||||
| Benchmarks | ||||
| MMLU-Pro | — | 88.2% | — | — |
| GPQA-D | — | 90.1% | — | — |
| HumanEval | — | 97.4% | — | — |
| Aider | — | 91.2% | — | — |
| AIME-25 | 79.8% | 93.5% | — | — |
| LiveCB | — | 80.2% | — | — |
| MMMU | — | 79.0% | — | — |
Add a model
Claude 4.8 OpusmaxQwen3.7-MaxmaxGemini 3.5maxGemini Omnimax
GPT-5.5maxQwen3.6-27BmaxDeepSeek-V4-FlashmaxDeepSeek-V4-PromaxGemini 3 PromaxClaude 4.7 Opus✓Claude 4.6 SonnetmaxClaude 4.5 HaikumaxDeepSeek-V3.1maxGPT-5max
Grok 4maxGemini 2.5 Flash✓Qwen3 235B✓o3maxGPT-4.1max
Llama 4 ScoutmaxLlama 4 MaverickmaxGemini 2.5 PromaxDeepSeek-R1✓
Mistral Large 2maxMax 4 models. Remove one to add another.