[whatmodel.sh](../../README.md) / StarCoderBase 3B
# StarCoderBase 3B
BigCode · released 2023-06 · 3B params · bigcode-openrail-m
**score 29.6** ±0.8 · #665 of 732 · from 12 benchmark results across 3 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks StarCoderBase 3B among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁●▃▅▇█▇▅▄▃▁ #451 of 469 96% 9
math ▁▁●▅█▆▄█▄▂▁▁ #402 of 431* 3.4% 1
coding · #451 of 469 · 96% of its score
Beats [Molmo 7B-D (0924)](molmo-7b-d.md), [Tulu 2 7B](tulu-2-7b.md), [DeepSeek R1 Distill Qwen 1.5B](deepseek-r1-distill-qwen-1-5b.md) here, though they rank higher overall. Trails [Falcon3 1B Instruct](falcon3-1b.md), [Granite 3.0 1B-A400M Instruct](granite-3-0-1b-a400m.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HumanEval 17.7% ▁·▁●▃▄▆▅█▃▁▁ beats 6/99 easy · avg 76% 41% sharp · several runs
HumanEval+ 15.9% ▁·▁▁●▃▄▄▄█▃▁ beats 6/99 easy · avg 70% 21% sharp
MBPP 29.4% ●·▅▃▃▃▅▃▃▅·▅ beats 0/15 easy · avg 81% 12% at its level · several runs
HumanEvalSynthesize (HumanEvalPac… 21.8% ▁·▂●▂▆▅█▄▄▁▂ beats 3/35 medium · avg 62% 7.0% typical
DS-1000 (maker) 14.2% ●▄·▄██▄█·█·▄ beats 0/12 medium · avg 49% 4.0% typical
CRUXEval-I 27.5% ●▅█▃▃▅·▃▃▅·▅ beats 0/15 medium · avg 56% 3.6% at its level
HumanEvalExplain (HumanEvalPack, … 11.9% ●▇█▃▄▇▄▁▇▃·▁ beats 2/35 medium · avg 46% 3.0% typical
CRUXEval-O 27.0% ●·▃██▅▃·▃▃▅▃ beats 0/15 medium · avg 54% 3.0% blunt · at its level
HumanEvalFix (HumanEvalPack, 6-la… 6.4% ▁·▁●▄██▄█▁▅▅ beats 2/35 medium · avg 51% 2.1% sharp · too hard for it
math · #402 of 431 · 3.4% of its score
Beats [Command R (03-2024)](command-r-03-2024.md), [Mistral 7B Instruct v0.3](mistral-7b-instruct-v0-3.md), [Llama-3-Instruct-8B-SimPO-v0.2](llama-3-instruct-8b-simpo-v0-2.md) here, though they rank higher overall. Trails [GPT-Neo 125M](gpt-neo-125m.md), [OPT-125M](opt-125m.md), [GPT-2 XL (1.5B)](gpt-2-xl.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K (maker) 8.0% ●▄▄·▄█·██·█▄ beats 0/12 easy · avg 75% 3.4% sharp
source results share of the score
EvalPlus / EvalPerf published evaluatio… 2 51% ▇▇▇▇▇▇
StarCoder 2 and The Stack v2: The Next … 4 26% ▇▇▇
Granite Code Models: A Family of Open F… 6 23% ▇▇▇
prior: pull toward the middle 0% ▇
weighted average = 29.6
Licences: EvalPlus / EvalPerf published evaluation runs ([Apache-2.0](../../licenses/evalplus-published-runs-Apache-2.0.txt)) · StarCoder 2 and The Stack v2: The Next Generation ([CC BY-SA 4.0](../../licenses/paper-2402.19173v1-CC-BY-SA-4.0.txt)) · Granite Code Models: A Family of Open Foundation Models for Code Intelligence ([CC BY 4.0](../../licenses/paper-2405.04324v1-CC-BY-4.0.txt))
## How to run it
### Ollama (recommended)
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run StarCoderBase 3B (Q4_0, needs ~1.8 GB, from [ollama.com](https://ollama.com/library/starcoder/tags))
```sh
ollama run starcoder:3b
```
Other versions: Q4_0 (~1.8 GB) `ollama run starcoder:3b`
### Other ways to run it
1. Run it as published (fp32, needs ~12.17 GB, from [huggingface.co/bigcode](https://huggingface.co/bigcode/starcoderbase-3b/tree/e1c5ef4ebb97afa0db09ec3e520f0487ca350bbe))
```sh
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='bigcode/starcoderbase-3b', torch_dtype='auto', device_map='auto')('def print_hello_world():', max_new_tokens=40)[0]['generated_text'])"
```
Model repository and documentation: [huggingface.co/bigcode](https://huggingface.co/bigcode/starcoderbase-3b)
[interactive version](https://whatmodel.sh/model/starcoderbase-3b) · [JSON](https://whatmodel.sh/api?format=json&all=1)