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 4.6 Sonnet×

Claude 4.7 Opus×

o3×

DeepSeek-V4-Flash×Clear all

	Claude 4.6 Sonnet Anthropic	Claude 4.7 Opus Anthropic	o3 OpenAI	DeepSeek-V4-Flash DeepSeek
Vendor	Anthropic	Anthropic	OpenAI	DeepSeek
Family	Claude	Claude	o-series	DeepSeek
Release date	2026-02-17	2026-04-16	2025-04-16	2026-04-22
Context window	1,000,000 tokens	1,000,000 tokens	200,000 tokens	1,000,000 tokens
Parameters	—	—	—	284B (13B active)
Modality	text, vision	text, vision	text, vision	text
License	proprietary	proprietary	proprietary	MIT
Source	proprietary	proprietary	proprietary	open weights
Description	Anthropic's mid-tier flagship — released 12 days after Opus 4.6 and matching most of its capabilities at ~40% the API cost ($3/$15 per 1M tokens). 1M-token context.	Anthropic's previous Opus flagship — superseded by Opus 4.8 on May 28, 2026 (42-day cycle).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.	Reasoning-focused model in the o-series.	Smaller, faster sibling to DeepSeek-V4-Pro. Same 1M context window with a much lighter 284B / 13B-active MoE.
Links		vendor →		weights →
Benchmarks
MMLU-Pro	85.4%	88.2%	—	—
GPQA-D	85.8%	90.1%	87.7%	—
HumanEval	95.9%	97.4%	—	—
Aider	83.6%	91.2%	—	—
AIME-25	87.4%	93.5%	88.9%	—
LiveCB	73.8%	80.2%	—	—
MMMU	—	79.0%	—	—