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

GPT-5 Pro×

Llama 4 Maverick×

Gemini 3.5×

DeepSeek-V3.1×Clear all

	GPT-5 Pro OpenAI	Llama 4 Maverick Meta	Gemini 3.5 Google	DeepSeek-V3.1 DeepSeek
Vendor	OpenAI	Meta	Google	DeepSeek
Family	GPT	Llama	Gemini	DeepSeek
Release date	2025-08-07	2025-04-05	2026-05-20	2025-08-21
Context window	400,000 tokens	1,000,000 tokens	2,000,000 tokens	128,000 tokens
Parameters	—	400B	—	671B
Modality	text, vision, audio	text, vision	text, vision, audio, video	text
License	proprietary	Llama 4 Community License	proprietary	MIT
Source	proprietary	open weights	proprietary	open weights
Description	Higher-reasoning variant of GPT-5, used for harder scientific and research tasks (e.g., immunology breakthroughs at NIH). Same 400K context as GPT-5; trades latency for depth.	Mixture-of-experts open-weight model from Meta.	Google DeepMind's next-gen Gemini — positioned as "frontier intelligence with action". Built for complex agentic workflows. Announced at Google I/O 2026.	Large MoE open-weight model. Predecessor to DeepSeek-V4.
Links	vendor →	weights →	vendor →	weights →
Benchmarks
MMLU-Pro	—	80.5%	88.9%	—
GPQA-D	—	—	91.2%	—
AIME-25	—	—	96.1%	—
MMMU	—	68.5%	86.0%	—