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

Claude 4.7 Opus×

DeepSeek-V4-Pro×

Gemini 3.5×

DeepSeek-R1×Clear all

	Claude 4.7 Opus Anthropic	DeepSeek-V4-Pro DeepSeek	Gemini 3.5 Google	DeepSeek-R1 DeepSeek
Vendor	Anthropic	DeepSeek	Google	DeepSeek
Family	Claude	DeepSeek	Gemini	DeepSeek
Release date	2026-04-16	2026-04-22	2026-05-20	2025-01-20
Context window	1,000,000 tokens	1,000,000 tokens	2,000,000 tokens	128,000 tokens
Parameters	—	1.6T (49B active)	—	671B
Modality	text, vision	text	text, vision, audio, video	text
License	proprietary	MIT	proprietary	MIT
Source	proprietary	open weights	proprietary	open weights
Description	Anthropic's previous Opus flagship — superseded by Opus 4.8 on May 28, 2026 (42-day cycle). Optimized for complex reasoning and coding. Improved software engineering, long-running coding tasks, and higher-resolution vision over Claude 4.6.	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.	Reasoning-focused open-weight model.
Links	vendor →	weights →	vendor →	weights →
Benchmarks
MMLU-Pro	88.2%	84.2%	88.9%	—
GPQA-D	90.1%	82.4%	91.2%	—
HumanEval	97.4%	95.1%	—	—
Aider	91.2%	80.1%	—	—
AIME-25	93.5%	88.6%	96.1%	79.8%
LiveCB	80.2%	72.4%	—	—
MMMU	79.0%	—	86.0%	—