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

Llama 4 Maverick×

Claude 4.6 Sonnet×

DeepSeek-V4-Pro×

DeepSeek-R1×Clear all

	Llama 4 Maverick Meta	Claude 4.6 Sonnet Anthropic	DeepSeek-V4-Pro DeepSeek	DeepSeek-R1 DeepSeek
Vendor	Meta	Anthropic	DeepSeek	DeepSeek
Family	Llama	Claude	DeepSeek	DeepSeek
Release date	2025-04-05	2026-02-17	2026-04-22	2025-01-20
Context window	1,000,000 tokens	1,000,000 tokens	1,000,000 tokens	128,000 tokens
Parameters	400B	—	1.6T (49B active)	671B
Modality	text, vision	text, vision	text	text
License	Llama 4 Community License	proprietary	MIT	MIT
Source	open weights	proprietary	open weights	open weights
Description	Mixture-of-experts open-weight model from Meta.	Anthropic's mid-tier flagship — released 12 days after Opus 4.6 and matching most of its capabilities at ~40% the API cost ($3/$15 per 1M tokens). 1M-token context.	DeepSeek's flagship open-weight MoE. 1.6T parameters with 49B activated, 1M-token context, and a hybrid attention scheme (CSA + HCA) that delivers long-context inference at ~27% of V3.2's FLOPs.	Reasoning-focused open-weight model.
Links	weights →		weights →	weights →
Benchmarks
MMLU-Pro	80.5%	85.4%	84.2%	—
GPQA-D	—	85.8%	82.4%	—
HumanEval	—	95.9%	95.1%	—
Aider	—	83.6%	80.1%	—
AIME-25	—	87.4%	88.6%	79.8%
LiveCB	—	73.8%	72.4%	—
MMMU	68.5%	—	—	—