[whatmodel.sh](../../README.md) / StarCoderBase 7B
# StarCoderBase 7B
BigCode · released 2023-07 · 7.53B params · bigcode-openrail-m
**score 32.7** ±0.7 · #646 of 732 · from 16 benchmark results across 4 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks StarCoderBase 7B among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁▁●▅▇█▇▅▄▃▁ #444 of 469 94% 9
math ▁▁●▅█▆▄█▄▂▁▁ #394 of 431 5.9% 2
coding · #444 of 469 · 94% of its score
Beats [Qwen3.5 2B](qwen3-5-2b.md), [Molmo 7B-D (0924)](molmo-7b-d.md), [Tulu 2 7B](tulu-2-7b.md) here, though they rank higher overall. Trails [Granite 3.0 1B-A400M Instruct](granite-3-0-1b-a400m.md), [Llama 3.2 1B Instruct](llama-3-2-1b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HumanEval 24.4% ▁·▁▁●▄▆▅█▃▁▁ beats 8/99 easy · avg 76% 40% sharp · at its level · several runs
HumanEval+ 21.3% ▁·▁▁▁●▄▄▄█▃▁ beats 10/99 easy · avg 70% 21% sharp
MBPP 34.8% █·●▃▃▃▅▃▃▅·▅ beats 3/15 easy · avg 81% 10% at its level · several runs
HumanEvalSynthesize (HumanEvalPac… 26.7% ▁·▂▁●▆▅█▄▄▁▂ beats 4/35 medium · avg 62% 6.3% at its level
CRUXEval-I 29.7% ▁··▁●▂█▇▁··▁ beats 3/22 medium · avg 56% 4.5% at its level · several runs
CRUXEval-O 32.2% ▁·▂●▁█▁▂···▁ beats 3/22 medium · avg 54% 4.0% blunt · at its level · several runs
DS-1000 (maker) 19.1% ▄▄·●██▄█·█·▄ beats 2/12 medium · avg 49% 3.6% typical
HumanEvalExplain (HumanEvalPack, … 14.8% ▄▇●▃▄▇▄▁▇▃·▁ beats 8/35 medium · avg 46% 3.0% typical
HumanEvalFix (HumanEvalPack, 6-la… 12.5% ▁·▁▁▄●█▄█▁▅▅ beats 9/35 medium · avg 51% 2.2% sharp
math · #394 of 431 · 5.9% of its score
Beats [Yi 34B Chat](yi-34b-chat.md), [Command R (03-2024)](command-r-03-2024.md), [DeepSeek LLM 67B Chat](deepseek-llm-67b-chat.md) here, though they rank higher overall. Trails [GPT-Neo 125M](gpt-neo-125m.md), [GPT-2 XL (1.5B)](gpt-2-xl.md), [GPT-Neo 1.3B](gpt-neo-1-3b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K (maker) 14.9% ▄▄●·▄█·██·█▄ beats 2/12 easy · avg 75% 4.9% sharp · several runs
MATH 2.4% ●·▃····▃···█ beats 0/5 medium · avg 37% 1.0% noisy
source results share of the score
EvalPlus / EvalPerf published evaluatio… 2 50% ▇▇▇▇▇▇
StarCoder 2 and The Stack v2: The Next … 4 24% ▇▇▇
Granite Code Models: A Family of Open F… 8 24% ▇▇▇
CRUXEval published evaluation results 2 3% ▇
prior: pull toward the middle 0% ▇
weighted average = 32.7
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)) · CRUXEval published evaluation results ([MIT](../../licenses/cruxeval-published-results-MIT.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 7B (Q4_0, needs ~4.3 GB, from [ollama.com](https://ollama.com/library/starcoder/tags))
```sh
ollama run starcoder:7b
```
Other versions: Q4_0 (~4.3 GB) `ollama run starcoder:7b`
### Other ways to run it
1. Run it as published (fp32, needs ~30.11 GB, from [huggingface.co/bigcode](https://huggingface.co/bigcode/starcoderbase-7b/tree/94e20e98e189b1d6dccd56e932667bd85cc7c393))
```sh
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='bigcode/starcoderbase-7b', 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-7b)
[interactive version](https://whatmodel.sh/model/starcoderbase-7b) · [JSON](https://whatmodel.sh/api?format=json&all=1)