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×

Llama 4 Maverick×

Grok 4×

GPT-5×Clear all

	Gemini 3 Pro Google	Llama 4 Maverick Meta	Grok 4 xAI	GPT-5 OpenAI
Vendor	Google	Meta	xAI	OpenAI
Family	Gemini	Llama	Grok	GPT
Release date	2026-04-22	2025-04-05	2025-07-09	2025-08-07
Context window	2,000,000 tokens	1,000,000 tokens	256,000 tokens	400,000 tokens
Parameters	—	400B	—	—
Modality	text, vision, audio, video	text, vision	text	text, vision, audio
License	proprietary	Llama 4 Community License	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.	Mixture-of-experts open-weight model from Meta.	—	OpenAI's flagship general-purpose model. Superseded by GPT-5.5 in April 2026.
Links	vendor →	weights →	vendor →	vendor →
Benchmarks
MMLU-Pro	87.1%	80.5%	82.8%	85.1%
GPQA-D	88.6%	—	87.5%	87.3%
HumanEval	96.5%	—	—	95.2%
Aider	84.8%	—	72.5%	85.4%
AIME-25	92.8%	—	86.1%	91.0%
LiveCB	76.1%	—	65.3%	74.5%
MMMU	84.3%	68.5%	—	74.1%