Compare models

Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.

Comparing agents instead? Switch to agent compare →

Selected

DeepSeek-R1×

Gemini 3.5×

Llama 4 Scout×

GPT-5×Clear all

	DeepSeek-R1 DeepSeek	Gemini 3.5 Google	Llama 4 Scout Meta	GPT-5 OpenAI
Vendor	DeepSeek	Google	Meta	OpenAI
Family	DeepSeek	Gemini	Llama	GPT
Release date	2025-01-20	2026-05-20	2025-04-05	2025-08-07
Context window	128,000 tokens	2,000,000 tokens	10,000,000 tokens	400,000 tokens
Parameters	671B	—	109B	—
Modality	text	text, vision, audio, video	text, vision	text, vision, audio
License	MIT	proprietary	Llama 4 Community License	proprietary
Source	open weights	proprietary	open weights	proprietary
Description	Reasoning-focused open-weight model.	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	—	OpenAI's flagship general-purpose model. Superseded by GPT-5.5 in April 2026.
Links	weights →	vendor →	weights →	vendor →
Benchmarks
MMLU-Pro	—	88.9%	—	85.1%
GPQA-D	—	91.2%	—	87.3%
HumanEval	—	—	—	95.2%
Aider	—	—	—	85.4%
AIME-25	79.8%	96.1%	—	91.0%
LiveCB	—	—	—	74.5%
MMMU	—	86.0%	—	74.1%