Compare · ModelsLive · 2 picked · head to head
Gemma 2B vs StarCoder 2 15B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemma 2B wins on 6/10 benchmarks
Gemma 2B wins 6 of 10 shared benchmarks. Leads in knowledge · general · reasoning.
Category leads
knowledge·Gemma 2Bmath·StarCoder 2 15Bgeneral·Gemma 2Blanguage·StarCoder 2 15Breasoning·Gemma 2B
Hype vs Reality
Attention vs performance
Gemma 2B
#189 by perf·no signal
StarCoder 2 15B
#204 by perf·no signal
Vendor risk
Who is behind the model
Google DeepMind
$4.00T·Tier 1
U
Unknown
private · undisclosed
Head to head
10 benchmarks · 2 models
Gemma 2BStarCoder 2 15B
ARC AI2
StarCoder 2 15B leads by +6.8
AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval.
Gemma 2B
22.8
StarCoder 2 15B
29.6
GSM8K
StarCoder 2 15B leads by +40.0
Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve.
Gemma 2B
17.7
StarCoder 2 15B
57.7
BBH (HuggingFace)
Gemma 2B leads by +0.8
Gemma 2B
21.1
StarCoder 2 15B
20.4
GPQA
Gemma 2B leads by +1.8
Gemma 2B
4.9
StarCoder 2 15B
3.1
IFEval
StarCoder 2 15B leads by +1.2
Gemma 2B
26.6
StarCoder 2 15B
27.8
MATH Level 5
Gemma 2B leads by +1.4
Gemma 2B
7.4
StarCoder 2 15B
6.0
MMLU-PRO
Gemma 2B leads by +6.6
Gemma 2B
21.6
StarCoder 2 15B
15.0
MUSR
Gemma 2B leads by +8.1
Gemma 2B
11.0
StarCoder 2 15B
2.9
MMLU
StarCoder 2 15B leads by +29.1
Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge.
Gemma 2B
23.1
StarCoder 2 15B
52.1
Winogrande
Gemma 2B leads by +2.2
WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs.
Gemma 2B
30.8
StarCoder 2 15B
28.6
Full benchmark table
| Benchmark | Gemma 2B | StarCoder 2 15B |
|---|---|---|
ARC AI2 AI2 Reasoning Challenge · tests grade-school level science knowledge with multiple-choice questions requiring reasoning beyond simple retrieval. | 22.8 | 29.6 |
GSM8K Grade School Math 8K · 8,500 linguistically diverse grade-school math word problems that require multi-step reasoning to solve. | 17.7 | 57.7 |
BBH (HuggingFace) | 21.1 | 20.4 |
GPQA | 4.9 | 3.1 |
IFEval | 26.6 | 27.8 |
MATH Level 5 | 7.4 | 6.0 |
MMLU-PRO | 21.6 | 15.0 |
MUSR | 11.0 | 2.9 |
MMLU Massive Multitask Language Understanding · 57 subjects spanning STEM, humanities, social sciences, and more. The standard benchmark for broad knowledge. | 23.1 | 52.1 |
Winogrande WinoGrande · large-scale commonsense reasoning benchmark where models must resolve ambiguous pronouns in carefully constructed sentence pairs. | 30.8 | 28.6 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| — | — | — | — | |
U StarCoder 2 15B | — | — | — | — |