The Model Index
Up to four models

Qwen3-30B-A3B vs DeepSeek-V3 (Mar 2025)

On the index (GPT-4 = 100), Qwen3-30B-A3B scores 125 and DeepSeek-V3 (Mar 2025) 124. 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 Qwen3-30B-A3B
Presets
Qwen3-30B-A3B
125109–141
WeightsOpen
ReleasedApr 2025
Price in / out$0.12 / $0.50
Context41K
Core tests taken1 of 9
DeepSeek-V3 (Mar 2025)
124108–140
WeightsOpen
ReleasedMar 2025
Price in / out$0.29 / $1.14
Context164K
Core tests taken2 of 9

Qwen3-30B-A3B and DeepSeek-V3 (Mar 2025) are too close to call (1 point apart, within the uncertainty), ahead on 0 of 1 shared core tests, and costs 2.3× less.

Test by testthird-party results

Qwen3-30B-A3BDeepSeek-V3 (Mar 2025)
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)Fiction.LiveBenchFiction.LiveBenchMATHSAIME-style maths (OTIS mock)AIME-style maths (OTIS mock)CODINGWeirdMLWeirdMLAGENTSDTBenchDTBenchLMCALMCA
Every result as a table
TestQwen3-30B-A3BDeepSeek-V3 (Mar 2025)
Reasoning & knowledge
GPQA Diamond (Epoch AI)61.7%67.6%
Fiction.LiveBench40.6%50.0%
Maths
AIME-style maths (OTIS mock)62.8%37.8%
Coding
WeirdML29.8%36.1%
Agents
DTBench60.3%64.8%
LMCA15.8%15.5%
Task length · METR30 sec

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

This page as a chartDownload PNG