Compare · ModelsLive · 2 picked · head to head
Qwen2.5 0.5B Instruct vs DeepSeek R1 Distill Qwen 1.5B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
DeepSeek R1 Distill Qwen 1.5B wins on 3/6 benchmarks
DeepSeek R1 Distill Qwen 1.5B wins 3 of 6 shared benchmarks. Leads in language · math · reasoning.
Category leads
general·Qwen2.5 0.5B Instructknowledge·Qwen2.5 0.5B Instructlanguage·DeepSeek R1 Distill Qwen 1.5Bmath·DeepSeek R1 Distill Qwen 1.5Breasoning·DeepSeek R1 Distill Qwen 1.5B
Hype vs Reality
Attention vs performance
Qwen2.5 0.5B Instruct
#229 by perf·no signal
DeepSeek R1 Distill Qwen 1.5B
#227 by perf·no signal
Best value
Pricing unknown
Qwen2.5 0.5B Instruct
—
no price
DeepSeek R1 Distill Qwen 1.5B
—
no price
Vendor risk
Mixed exposure
One or more vendors flagged
Alibaba (Qwen)
$293.0B·Tier 1
DeepSeek
$3.4B·Tier 1
Head to head
6 benchmarks · 2 models
Qwen2.5 0.5B InstructDeepSeek R1 Distill Qwen 1.5B
BBH (HuggingFace)
Qwen2.5 0.5B Instruct leads by +3.4
Qwen2.5 0.5B Instruct
8.2
DeepSeek R1 Distill Qwen 1.5B
4.7
GPQA
Qwen2.5 0.5B Instruct leads by +0.4
Qwen2.5 0.5B Instruct
1.2
DeepSeek R1 Distill Qwen 1.5B
0.8
IFEval
DeepSeek R1 Distill Qwen 1.5B leads by +3.1
Qwen2.5 0.5B Instruct
31.5
DeepSeek R1 Distill Qwen 1.5B
34.6
MATH Level 5
DeepSeek R1 Distill Qwen 1.5B leads by +6.6
Qwen2.5 0.5B Instruct
10.3
DeepSeek R1 Distill Qwen 1.5B
16.9
MMLU-PRO
Qwen2.5 0.5B Instruct leads by +5.9
Qwen2.5 0.5B Instruct
8.0
DeepSeek R1 Distill Qwen 1.5B
2.1
MUSR
DeepSeek R1 Distill Qwen 1.5B leads by +1.6
Qwen2.5 0.5B Instruct
1.4
DeepSeek R1 Distill Qwen 1.5B
3.0
Full benchmark table
| Benchmark | Qwen2.5 0.5B Instruct | DeepSeek R1 Distill Qwen 1.5B |
|---|---|---|
BBH (HuggingFace) | 8.2 | 4.7 |
GPQA | 1.2 | 0.8 |
IFEval | 31.5 | 34.6 |
MATH Level 5 | 10.3 | 16.9 |
MMLU-PRO | 8.0 | 2.1 |
MUSR | 1.4 | 3.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
| — | — | — | — |