The Model Index
Up to four models

gpt-oss-20b vs Qwen3-30B-A3B-Instruct (Jul 2025)

On the index (GPT-4 = 100), gpt-oss-20b scores 128 and Qwen3-30B-A3B-Instruct (Jul 2025) 127. 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 gpt-oss-20b
Presets
gpt-oss-20b
128112–144
WeightsOpen
ReleasedAug 2025
Price in / out$0.018 / $0.090
Context131K
Core tests taken1 of 9
Qwen3-30B-A3B-Instruct (Jul 2025)
127111–143
WeightsOpen
ReleasedJul 2025
Price in / out$0.10 / $0.30
Context262K
Core tests taken1 of 9

gpt-oss-20b and Qwen3-30B-A3B-Instruct (Jul 2025) are too close to call (1 point apart, within the uncertainty), ahead on 1 of 1 shared core tests, and costs 4.2× less.

Test by testthird-party results

gpt-oss-20bQwen3-30B-A3B-Instruct (Jul 2025)
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)MATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)AGENTSDTBenchDTBenchLMCALMCA
Every result as a table
Testgpt-oss-20bQwen3-30B-A3B-Instruct (Jul 2025)
Reasoning & knowledge
GPQA Diamond (Epoch AI)60.8%55.6%
Maths
AIME-style maths (OTIS mock)65.3%62.2%
Agents
DTBench68.0%67.2%
LMCA14.5%22.4%

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

This page as a chartDownload PNG