[whatmodel.sh](../../README.md) / StarCoder2 3B
# StarCoder2 3B
BigCode · released 2024-02 · 3.03B params · bigcode-openrail-m
**score 35.1** ±0.6 · #625 of 732 · from 17 benchmark results across 5 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks StarCoder2 3B among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁▁●▅▇█▇▅▄▃▁ #427 of 469 87% 9
math ▁▁▃●█▆▄█▄▂▁▁ #353 of 431* 2.8% 1
coding · #427 of 469 · 87% of its score
Beats [DeepSeek R1 Distill Qwen 7B](deepseek-r1-distill-qwen-7b.md), [Qwen3.5 2B](qwen3-5-2b.md), [DeepSeek R1 Distill Llama 8B](deepseek-r1-distill-llama-8b.md) here, though they rank higher overall. Trails [Llama 3.2 1B Instruct](llama-3-2-1b.md), [SmolLM2 1.7B Instruct](smollm2-1-7b.md), [Granite 3B Code Base (2K)](granite-3b-code-base-2k.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HumanEval 31.7% ▁·▁▁▃●▆▅█▃▁▁ beats 16/99 easy · avg 76% 46% sharp · at its level · several runs
HumanEval+ 27.4% ▁·▁▁▁●▄▄▄█▃▁ beats 16/99 easy · avg 70% 17% sharp · at its level
MBPP 42.4% █·▅▃▃▃●▃▃▅·▅ beats 9/15 easy · avg 81% 7.1% at its level · several runs
HumanEvalSynthesize (HumanEvalPac… 31.7% ▁·▂▁▂●▅█▄▄▁▂ beats 9/35 medium · avg 62% 4.9% at its level
CRUXEval-I 32.1% ▃▅●▃▃▅·▃▃▅·▅ beats 5/15 medium · avg 56% 3.0% at its level · several runs
DS-1000 (maker) 25.0% ▄▄·▄█●▄█·█·▄ beats 6/12 medium · avg 49% 2.9% typical
HumanEvalExplain (HumanEvalPack, … 13.1% ▄●█▃▄▇▄▁▇▃·▁ beats 5/35 medium · avg 46% 2.5% typical
CRUXEval-O 34.0% ▃·▃█●▅▃·▃▃▅▃ beats 6/15 medium · avg 54% 2.5% blunt · at its level · several runs
HumanEvalFix (HumanEvalPack, 6-la… 11.0% ▁·▁▁▄●█▄█▁▅▅ beats 8/35 medium · avg 51% 2.0% sharp
math · #353 of 431 · 2.8% of its score
Beats [GPT-3.5 Turbo (0125)](gpt-3-5-turbo-0125.md), [DBRX Instruct](dbrx-instruct.md), [Phi-3 Medium 128K Instruct](phi-3-medium-128k-instruct.md) here, though they rank higher overall. Trails [OPT-2.7B](opt-2-7b.md), [Gemma 2B (base)](gemma-2b-base.md), [Stable Code 3B](stable-code-3b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K (maker) 27.7% ▄▄▄·▄●·██·█▄ beats 5/12 easy · avg 75% 2.8% sharp · at its level
other tests (2)
test result vs the others in its run difficulty counts why
EvoEval (overall) 20.2% ▂▂●▄▄█▁▄▅▃▁▃ beats 4/42 medium · avg 52% 6.6% typical
BigCodeBench (Complete) (maker) 21.4% ●▁▁▄▂▅▇█▆▂▂▂ beats 1/58 medium · avg 48% 3.0% at its level
source results share of the score
EvalPlus / EvalPerf published evaluatio… 2 41% ▇▇▇▇▇
Top Leaderboard Ranking = Top Coding Pr… 2 20% ▇▇
StarCoder 2 and The Stack v2: The Next … 6 20% ▇▇
Granite Code Models: A Family of Open F… 6 17% ▇▇
BigCodeBench paper (submitted / maker-r… 1 3% ▇
prior: pull toward the middle 0% ▇
weighted average = 35.1
Licences: EvalPlus / EvalPerf published evaluation runs ([Apache-2.0](../../licenses/evalplus-published-runs-Apache-2.0.txt)) · Top Leaderboard Ranking = Top Coding Proficiency ([CC BY 4.0](../../licenses/paper-2403.19114v1-CC-BY-4.0.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)) · BigCodeBench paper ([CC BY 4.0](http://creativecommons.org/licenses/by/4.0/))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run StarCoder2 3B (Q4_K_M, needs ~1.8 GB, from [huggingface.co/second-state](https://huggingface.co/second-state/StarCoder2-3B-GGUF/tree/7fca3e2da2ce31df411461e2cb9cae2d2b492f35))
```sh
llama cli -hf second-state/StarCoder2-3B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf second-state/StarCoder2-3B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~3.2 GB) `llama cli -hf second-state/StarCoder2-3B-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run StarCoder2 3B (Q4_K_M, needs ~1.8 GB, from [huggingface.co/second-state](https://huggingface.co/second-state/StarCoder2-3B-GGUF/tree/7fca3e2da2ce31df411461e2cb9cae2d2b492f35))
```sh
lms get https://huggingface.co/second-state/StarCoder2-3B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~3.2 GB) `lms get https://huggingface.co/second-state/StarCoder2-3B-GGUF@Q8_0 -y` · 4bit (~1.9 GB, Mac only) `lms get https://huggingface.co/mlx-community/starcoder2-3b-4bit -y` · 8bit (~3.4 GB, Mac only) `lms get https://huggingface.co/OmAr-Kader-DeV/starcoder2-3b-mlx-8Bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run StarCoder2 3B (Q4_K_M, needs ~1.8 GB, from [ollama.com](https://ollama.com/library/starcoder2/tags))
```sh
ollama run starcoder2:3b-q4_K_M
```
Other versions: Q8_0 (~3.2 GB) `ollama run starcoder2:3b-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run StarCoder2 3B (4bit, needs ~1.9 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/starcoder2-3b-4bit/tree/d90b61f0a26e018c1505ea6ed0fdfeca4e649789))
```sh
mlx_lm.chat --model mlx-community/starcoder2-3b-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/starcoder2-3b-4bit
```
Other versions: 8bit (~3.4 GB, Mac only) `mlx_lm.chat --model OmAr-Kader-DeV/starcoder2-3b-mlx-8Bit`
Model repository and documentation: [huggingface.co/bigcode](https://huggingface.co/bigcode/starcoder2-3b)
[interactive version](https://whatmodel.sh/model/starcoder2-3b) · [JSON](https://whatmodel.sh/api?format=json&all=1)