Compare models
Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.
 DeepSeek-R1DeepSeek |  Gemini 3 Pro | Gemini 2.5 Pro |  Llama 4 ScoutMeta | |
|---|---|---|---|---|
| Vendor | DeepSeek | Meta | ||
| Family | DeepSeek | Gemini | Gemini | Llama |
| Release date | 2025-01-20 | 2026-04-22 | 2025-03-25 | 2025-04-05 |
| Context window | 128,000 tokens | 2,000,000 tokens | 2,000,000 tokens | 10,000,000 tokens |
| Parameters | 671B | — | — | 109B |
| Modality | text | text, vision, audio, video | text, vision, audio, video | text, vision |
| License | MIT | proprietary | proprietary | Llama 4 Community License |
| Source | open weights | proprietary | proprietary | open weights |
| Description | Reasoning-focused open-weight model. | Google DeepMind's flagship multimodal model. Best-in-class for multimodal understanding and agentic / vibe coding workflows. A "Deep Think" reasoning mode is available to AI Ultra subscribers. | Predecessor to Gemini 3 Pro. | — |
| Links |
Add a model
Gemini 2.5 Flashmax
Claude 4.6 Sonnetmax
GPT-5.5maxDeepSeek-V4-Flashmax
Qwen3.6-27BmaxGemini 3 Pro✓DeepSeek-V4-PromaxClaude 4.7 OpusmaxClaude 4.5 HaikumaxDeepSeek-V3.1maxGPT-5max
Grok 4maxQwen3 235Bmaxo3maxGPT-4.1maxLlama 4 Scout✓Llama 4 MaverickmaxGemini 2.5 Pro✓DeepSeek-R1✓
Mistral Large 2maxMax 4 models. Remove one to add another.