Compare · ModelsLive · 2 picked · head to head
MiniMax M2 vs Qwen3 30B A3B Thinking 2507
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
MiniMax M2 wins on 5/6 benchmarks
MiniMax M2 wins 5 of 6 shared benchmarks. Leads in knowledge · language · coding.
Category leads
math·Qwen3 30B A3B Thinking 2507knowledge·MiniMax M2language·MiniMax M2coding·MiniMax M2
Hype vs Reality
Attention vs performance
MiniMax M2
#20 by perf·#11 by attention
Qwen3 30B A3B Thinking 2507
#32 by perf·#2 by attention
Best value
MiniMax M2
1.8x better value than Qwen3 30B A3B Thinking 2507
MiniMax M2
92.7 pts/$
$0.75/M
Qwen3 30B A3B Thinking 2507
50.9 pts/$
$1.30/M
Vendor risk
Mixed exposure
One or more vendors flagged
MiniMax
$4.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
6 benchmarks · 2 models
MiniMax M2Qwen3 30B A3B Thinking 2507
OpenCompass · AIME2025
Qwen3 30B A3B Thinking 2507 leads by +7.7
MiniMax M2
79.1
Qwen3 30B A3B Thinking 2507
86.8
OpenCompass · GPQA-Diamond
MiniMax M2 leads by +8.6
MiniMax M2
78.7
Qwen3 30B A3B Thinking 2507
70.1
OpenCompass · HLE
MiniMax M2 leads by +1.7
MiniMax M2
13.4
Qwen3 30B A3B Thinking 2507
11.7
OpenCompass · IFEval
MiniMax M2 leads by +0.5
MiniMax M2
90.2
Qwen3 30B A3B Thinking 2507
89.7
OpenCompass · LiveCodeBenchV6
MiniMax M2 leads by +14.6
MiniMax M2
74.0
Qwen3 30B A3B Thinking 2507
59.4
OpenCompass · MMLU-Pro
MiniMax M2 leads by +2.1
MiniMax M2
81.6
Qwen3 30B A3B Thinking 2507
79.5
Full benchmark table
| Benchmark | MiniMax M2 | Qwen3 30B A3B Thinking 2507 |
|---|---|---|
OpenCompass · AIME2025 | 79.1 | 86.8 |
OpenCompass · GPQA-Diamond | 78.7 | 70.1 |
OpenCompass · HLE | 13.4 | 11.7 |
OpenCompass · IFEval | 90.2 | 89.7 |
OpenCompass · LiveCodeBenchV6 | 74.0 | 59.4 |
OpenCompass · MMLU-Pro | 81.6 | 79.5 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.30 | $1.20 | 205K tokens (~102 books) | $5.25 | |
| $0.20 | $2.40 | 82K tokens (~41 books) | $7.50 |