The Model Index
Up to four models

Qwen3-235B-A22B-Instruct (Jul 2025) vs DeepSeek-R1

On the index (GPT-4 = 100), Qwen3-235B-A22B-Instruct (Jul 2025) scores 130 and DeepSeek-R1 130. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Qwen3-235B-A22B-Instruct (Jul 2025)
Presets
Qwen3-235B-A22B-Instruct (Jul 2025)
130114–146
WeightsOpen
ReleasedJul 2025
Price in / out$0.14 / $0.80
Context262K
Core tests taken0 of 9
DeepSeek-R1
130114–146
WeightsOpen
ReleasedJan 2025
Price in / out–
Context–
Core tests taken2 of 9

Qwen3-235B-A22B-Instruct (Jul 2025) and DeepSeek-R1 are too close to call (0 points apart, within the uncertainty).

Test by testthird-party results

Qwen3-235B-A22B-Instruct (Jul 2025)DeepSeek-R1
0%25%50%75%100%REASONING & KNOWLEDGEFiction.LiveBenchFiction.LiveBenchCODINGAider PolyglotAider PolyglotWeirdMLWeirdML
Every result as a table
TestQwen3-235B-A22B-Instruct (Jul 2025)DeepSeek-R1
Reasoning & knowledge
Fiction.LiveBench52.9%69.4%
Coding
Aider Polyglot59.6%56.9%
WeirdML38.7%36.5%
Task length · METR31 sec

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG