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

Grok 4×

Llama 4 Scout×

Gemini 3.5×

Gemini 3 Pro×Clear all

	Grok 4 xAI	Llama 4 Scout Meta	Gemini 3.5 Google	Gemini 3 Pro Google
Vendor	xAI	Meta	Google	Google
Family	Grok	Llama	Gemini	Gemini
Release date	2025-07-09	2025-04-05	2026-05-20	2026-04-22
Context window	256,000 tokens	10,000,000 tokens	2,000,000 tokens	2,000,000 tokens
Parameters	—	109B	—	—
Modality	text	text, vision	text, vision, audio, video	text, vision, audio, video
License	proprietary	Llama 4 Community License	proprietary	proprietary
Source	proprietary	open weights	proprietary	proprietary
Description	—	—	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	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.
Links	vendor →	weights →	vendor →	vendor →
Benchmarks
MMLU-Pro	82.8%	—	88.9%	87.1%
GPQA-D	87.5%	—	91.2%	88.6%
HumanEval	—	—	—	96.5%
Aider	72.5%	—	—	84.8%
AIME-25	86.1%	—	96.1%	92.8%
LiveCB	65.3%	—	—	76.1%
MMMU	—	—	86.0%	84.3%