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

Qwen3.6-27B×

Grok 4×

Claude 4.6 Sonnet×

DeepSeek-V4-Pro×Clear all

	Qwen3.6-27B Alibaba	Grok 4 xAI	Claude 4.6 Sonnet Anthropic	DeepSeek-V4-Pro DeepSeek
Vendor	Alibaba	xAI	Anthropic	DeepSeek
Family	Qwen	Grok	Claude	DeepSeek
Release date	2026-04-22	2025-07-09	2026-02-17	2026-04-22
Context window	262,144 tokens	256,000 tokens	1,000,000 tokens	1,000,000 tokens
Parameters	27B (dense)	—	—	1.6T (49B active)
Modality	text, vision, video	text	text, vision	text
License	Apache-2.0	proprietary	proprietary	MIT
Source	open weights	proprietary	proprietary	open weights
Description	Alibaba's first dense open-weight in the 3.6 family. Strong agentic-coding scores (77.2 SWE-bench Verified, matching Claude 4.5 Opus on Terminal-Bench 2.0). Supports 201 languages and multimodal text/image/video input.	—	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.	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.
Links	weights →	vendor →		weights →
Benchmarks
MMLU-Pro	78.9%	82.8%	85.4%	84.2%
GPQA-D	—	87.5%	85.8%	82.4%
HumanEval	91.5%	—	95.9%	95.1%
Aider	70.2%	72.5%	83.6%	80.1%
AIME-25	—	86.1%	87.4%	88.6%
LiveCB	—	65.3%	73.8%	72.4%
MMMU	70.3%	—	—	—