Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
Comparing agents instead? Switch to agent compare →
 Grok 4xAI |  GPT-5.5OpenAI |  Qwen3.7-MaxAlibaba |  Gemini 2.5 Pro | |
|---|---|---|---|---|
| Vendor | xAI | OpenAI | Alibaba | |
| Family | Grok | GPT | Qwen | Gemini |
| Release date | 2025-07-09 | 2026-04-23 | 2026-05-20 | 2025-03-25 |
| Context window | 256,000 tokens | 400,000 tokens | 1,000,000 tokens | 2,000,000 tokens |
| Parameters | — | — | — | — |
| Modality | text | text, vision, audio | text | text, vision, audio, video |
| License | proprietary | proprietary | proprietary | proprietary |
| Source | proprietary | proprietary | proprietary | proprietary |
| Description | — | OpenAI's smartest and most intuitive model — successor to GPT-5, with major coding, research, and document workflow gains. GPT-5.5 Pro variant also available for heavier reasoning. | Alibaba's flagship agent model — 1M-token context, extended-thinking mode, 56.6 on the Artificial Analysis Intelligence Index v4.0 (5th overall, #1 Chinese). 50.8% on Terminal-Bench Hard. Designed for long-horizon agent workloads (hundreds-to-thousands of steps). Closed-weight, $2.50/$7.50 per 1M tokens. | Predecessor to Gemini 3 Pro. |
| Links | ||||
| Benchmarks | ||||
| MMLU-Pro | 82.8% | 87.6% | 83.7% | 84.1% |
| GPQA-D | 87.5% | 89.4% | 83.0% | 84.0% |
| HumanEval | — | 96.8% | 93.9% | — |
| Aider | 72.5% | 89.7% | 78.4% | — |
| AIME-25 | 86.1% | 94.2% | 90.4% | — |
| LiveCB | 65.3% | 78.9% | 71.0% | — |
| MMMU | — | 78.4% | — | 72.4% |
Add a model
Claude 4.8 OpusmaxQwen3.7-Max✓Gemini 3.5maxGemini OmnimaxGPT-5.5✓Qwen3.6-27Bmax
DeepSeek-V4-FlashmaxDeepSeek-V4-PromaxGemini 3 PromaxClaude 4.7 OpusmaxClaude 4.6 SonnetmaxClaude 4.5 HaikumaxDeepSeek-V3.1maxGPT-5maxGrok 4✓Gemini 2.5 FlashmaxQwen3 235Bmaxo3maxGPT-4.1max
Llama 4 ScoutmaxLlama 4 MaverickmaxGemini 2.5 Pro✓DeepSeek-R1max
Mistral Large 2maxMax 4 models. Remove one to add another.