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×

Grok 4×

Llama 4 Maverick×

Claude 4.7 Opus×Clear all

	DeepSeek-R1 DeepSeek	Grok 4 xAI	Llama 4 Maverick Meta	Claude 4.7 Opus Anthropic
Vendor	DeepSeek	xAI	Meta	Anthropic
Family	DeepSeek	Grok	Llama	Claude
Release date	2025-01-20	2025-07-09	2025-04-05	2026-04-16
Context window	128,000 tokens	256,000 tokens	1,000,000 tokens	1,000,000 tokens
Parameters	671B	—	400B	—
Modality	text	text	text, vision	text, vision
License	MIT	proprietary	Llama 4 Community License	proprietary
Source	open weights	proprietary	open weights	proprietary
Description	Reasoning-focused open-weight model.	—	Mixture-of-experts open-weight model from Meta.	Anthropic's previous Opus flagship — superseded by Opus 4.8 on May 28, 2026 (42-day cycle).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.
Links	weights →	vendor →	weights →	vendor →
Benchmarks
MMLU-Pro	—	82.8%	80.5%	88.2%
GPQA-D	—	87.5%	—	90.1%
HumanEval	—	—	—	97.4%
Aider	—	72.5%	—	91.2%
AIME-25	79.8%	86.1%	—	93.5%
LiveCB	—	65.3%	—	80.2%
MMMU	—	—	68.5%	79.0%