Compare · ModelsLive · 2 picked · head to head
Qwen2.5 Coder 32B Instruct vs Qwen2.5 14B Instruct
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen2.5 Coder 32B Instruct wins on 3/6 benchmarks
Qwen2.5 Coder 32B Instruct wins 3 of 6 shared benchmarks. Leads in general · knowledge · reasoning.
Category leads
general·Qwen2.5 Coder 32B Instructknowledge·Qwen2.5 Coder 32B Instructlanguage·Qwen2.5 14B Instructmath·Qwen2.5 14B Instructreasoning·Qwen2.5 Coder 32B Instruct
Hype vs Reality
Attention vs performance
Qwen2.5 Coder 32B Instruct
#83 by perf·no signal
Qwen2.5 14B Instruct
#135 by perf·no signal
Best value
Qwen2.5 Coder 32B Instruct
Qwen2.5 Coder 32B Instruct
64.0 pts/$
$0.83/M
Qwen2.5 14B Instruct
—
no price
Vendor risk
Who is behind the model
Alibaba (Qwen)
$293.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
6 benchmarks · 2 models
Qwen2.5 Coder 32B InstructQwen2.5 14B Instruct
BBH (HuggingFace)
Qwen2.5 Coder 32B Instruct leads by +3.7
Qwen2.5 Coder 32B Instruct
52.3
Qwen2.5 14B Instruct
48.6
GPQA
Qwen2.5 Coder 32B Instruct leads by +2.7
Qwen2.5 Coder 32B Instruct
13.2
Qwen2.5 14B Instruct
10.5
IFEval
Qwen2.5 14B Instruct leads by +8.8
Qwen2.5 Coder 32B Instruct
72.7
Qwen2.5 14B Instruct
81.4
MATH Level 5
Qwen2.5 14B Instruct leads by +5.7
Qwen2.5 Coder 32B Instruct
49.5
Qwen2.5 14B Instruct
55.3
MMLU-PRO
Qwen2.5 14B Instruct leads by +5.3
Qwen2.5 Coder 32B Instruct
37.9
Qwen2.5 14B Instruct
43.2
MUSR
Qwen2.5 Coder 32B Instruct leads by +3.1
Qwen2.5 Coder 32B Instruct
13.7
Qwen2.5 14B Instruct
10.6
Full benchmark table
| Benchmark | Qwen2.5 Coder 32B Instruct | Qwen2.5 14B Instruct |
|---|---|---|
BBH (HuggingFace) | 52.3 | 48.6 |
GPQA | 13.2 | 10.5 |
IFEval | 72.7 | 81.4 |
MATH Level 5 | 49.5 | 55.3 |
MMLU-PRO | 37.9 | 43.2 |
MUSR | 13.7 | 10.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.66 | $1.00 | 33K tokens (~16 books) | $7.45 | |
| — | — | — | — |
People also compared