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

DeepSeek-V4-Pro×

GPT-5.5×

Gemini 3.5×

DeepSeek-V3.1×Clear all

	DeepSeek-V4-Pro DeepSeek	GPT-5.5 OpenAI	Gemini 3.5 Google	DeepSeek-V3.1 DeepSeek
Vendor	DeepSeek	OpenAI	Google	DeepSeek
Family	DeepSeek	GPT	Gemini	DeepSeek
Release date	2026-04-22	2026-04-23	2026-05-20	2025-08-21
Context window	1,000,000 tokens	400,000 tokens	2,000,000 tokens	128,000 tokens
Parameters	1.6T (49B active)	—	—	671B
Modality	text	text, vision, audio	text, vision, audio, video	text
License	MIT	proprietary	proprietary	MIT
Source	open weights	proprietary	proprietary	open weights
Description	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.	OpenAI's smartest and most intuitive model — successor to GPT-5, with major coding, research, and document workflow gains. GPT-5.5 Pro variant also available for heavier reasoning.	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	weights →	vendor →	vendor →	weights →
Benchmarks
MMLU-Pro	84.2%	87.6%	88.9%	—
GPQA-D	82.4%	89.4%	91.2%	—
HumanEval	95.1%	96.8%	—	—
Aider	80.1%	89.7%	—	—
AIME-25	88.6%	94.2%	96.1%	—
LiveCB	72.4%	78.9%	—	—
MMMU	—	78.4%	86.0%	—