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

Gemini 3 Pro×

DeepSeek-R1×

Grok 4×

Claude 4.6 Sonnet×Clear all

	Gemini 3 Pro Google	DeepSeek-R1 DeepSeek	Grok 4 xAI	Claude 4.6 Sonnet Anthropic
Vendor	Google	DeepSeek	xAI	Anthropic
Family	Gemini	DeepSeek	Grok	Claude
Release date	2026-04-22	2025-01-20	2025-07-09	2026-02-17
Context window	2,000,000 tokens	128,000 tokens	256,000 tokens	1,000,000 tokens
Parameters	—	671B	—	—
Modality	text, vision, audio, video	text	text	text, vision
License	proprietary	MIT	proprietary	proprietary
Source	proprietary	open weights	proprietary	proprietary
Description	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.	Reasoning-focused open-weight model.	—	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.
Links	vendor →	weights →	vendor →
Benchmarks
MMLU-Pro	87.1%	—	82.8%	85.4%
GPQA-D	88.6%	—	87.5%	85.8%
HumanEval	96.5%	—	—	95.9%
Aider	84.8%	—	72.5%	83.6%
AIME-25	92.8%	79.8%	86.1%	87.4%
LiveCB	76.1%	—	65.3%	73.8%
MMMU	84.3%	—	—	—