Compare · ModelsLive · 2 picked · head to head
Qwen2 0.5B Instruct vs SmolLM2 135M Instruct
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen2 0.5B Instruct wins on 4/6 benchmarks
Qwen2 0.5B Instruct wins 4 of 6 shared benchmarks. Leads in general · knowledge · math.
Category leads
general·Qwen2 0.5B Instructknowledge·Qwen2 0.5B Instructlanguage·SmolLM2 135M Instructmath·Qwen2 0.5B Instructreasoning·SmolLM2 135M Instruct
Hype vs Reality
Attention vs performance
Qwen2 0.5B Instruct
#232 by perf·no signal
SmolLM2 135M Instruct
#220 by perf·no signal
Best value
Pricing unknown
Qwen2 0.5B Instruct
—
no price
SmolLM2 135M Instruct
—
no price
Vendor risk
Who is behind the model
Alibaba (Qwen)
$293.0B·Tier 1
HF
Hugging Face TB
private · undisclosed
Head to head
6 benchmarks · 2 models
Qwen2 0.5B InstructSmolLM2 135M Instruct
BBH (HuggingFace)
Qwen2 0.5B Instruct leads by +1.2
Qwen2 0.5B Instruct
5.9
SmolLM2 135M Instruct
4.7
GPQA
Qwen2 0.5B Instruct
0.0
SmolLM2 135M Instruct
0.0
IFEval
SmolLM2 135M Instruct leads by +6.4
Qwen2 0.5B Instruct
22.5
SmolLM2 135M Instruct
28.8
MATH Level 5
Qwen2 0.5B Instruct leads by +2.6
Qwen2 0.5B Instruct
2.9
SmolLM2 135M Instruct
0.3
MMLU-PRO
Qwen2 0.5B Instruct leads by +4.6
Qwen2 0.5B Instruct
5.9
SmolLM2 135M Instruct
1.3
MUSR
SmolLM2 135M Instruct leads by +1.3
Qwen2 0.5B Instruct
2.4
SmolLM2 135M Instruct
3.7
Full benchmark table
| Benchmark | Qwen2 0.5B Instruct | SmolLM2 135M Instruct |
|---|---|---|
BBH (HuggingFace) | 5.9 | 4.7 |
GPQA | 0.0 | 0.0 |
IFEval | 22.5 | 28.8 |
MATH Level 5 | 2.9 | 0.3 |
MMLU-PRO | 5.9 | 1.3 |
MUSR | 2.4 | 3.7 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| Qwen2 0.5B Instruct | — | — | — | — |
HF SmolLM2 135M Instruct | — | — | — | — |