Compare · ModelsLive · 2 picked · head to head
Qwen2.5 1.5B Instruct vs MPT-30B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen2.5 1.5B Instruct wins on 5/6 benchmarks
Qwen2.5 1.5B Instruct wins 5 of 6 shared benchmarks. Leads in general · language · math.
Category leads
general·Qwen2.5 1.5B Instructknowledge·MPT-30Blanguage·Qwen2.5 1.5B Instructmath·Qwen2.5 1.5B Instructreasoning·Qwen2.5 1.5B Instruct
Hype vs Reality
Attention vs performance
Qwen2.5 1.5B Instruct
#215 by perf·no signal
MPT-30B
#180 by perf·no signal
Vendor risk
Who is behind the model
Alibaba (Qwen)
$293.0B·Tier 1
U
Unknown
private · undisclosed
Head to head
6 benchmarks · 2 models
Qwen2.5 1.5B InstructMPT-30B
BBH (HuggingFace)
Qwen2.5 1.5B Instruct leads by +13.3
Qwen2.5 1.5B Instruct
19.8
MPT-30B
6.5
GPQA
MPT-30B leads by +0.6
Qwen2.5 1.5B Instruct
0.8
MPT-30B
1.3
IFEval
Qwen2.5 1.5B Instruct leads by +23.2
Qwen2.5 1.5B Instruct
44.8
MPT-30B
21.5
MATH Level 5
Qwen2.5 1.5B Instruct leads by +20.5
Qwen2.5 1.5B Instruct
22.1
MPT-30B
1.6
MMLU-PRO
Qwen2.5 1.5B Instruct leads by +17.7
Qwen2.5 1.5B Instruct
20.0
MPT-30B
2.3
MUSR
Qwen2.5 1.5B Instruct leads by +0.3
Qwen2.5 1.5B Instruct
3.2
MPT-30B
2.9
Full benchmark table
| Benchmark | Qwen2.5 1.5B Instruct | MPT-30B |
|---|---|---|
BBH (HuggingFace) | 19.8 | 6.5 |
GPQA | 0.8 | 1.3 |
IFEval | 44.8 | 21.5 |
MATH Level 5 | 22.1 | 1.6 |
MMLU-PRO | 20.0 | 2.3 |
MUSR | 3.2 | 2.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| Qwen2.5 1.5B Instruct | — | — | — | — |
U MPT-30B | — | — | — | — |