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-V4-Pro×

Gemini 3.5×

Grok 4×Clear all

	Gemini 3 Pro Google	DeepSeek-V4-Pro DeepSeek	Gemini 3.5 Google	Grok 4 xAI
Vendor	Google	DeepSeek	Google	xAI
Family	Gemini	DeepSeek	Gemini	Grok
Release date	2026-04-22	2026-04-22	2026-05-20	2025-07-09
Context window	2,000,000 tokens	1,000,000 tokens	2,000,000 tokens	256,000 tokens
Parameters	—	1.6T (49B active)	—	—
Modality	text, vision, audio, video	text	text, vision, audio, video	text
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.	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.	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	—
Links	vendor →	weights →	vendor →	vendor →
Benchmarks
MMLU-Pro	87.1%	84.2%	88.9%	82.8%
GPQA-D	88.6%	82.4%	91.2%	87.5%
HumanEval	96.5%	95.1%	—	—
Aider	84.8%	80.1%	—	72.5%
AIME-25	92.8%	88.6%	96.1%	86.1%
LiveCB	76.1%	72.4%	—	65.3%
MMMU	84.3%	—	86.0%	—