whatmodel.sh / StarCoder2 15B

StarCoder2 15B

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

StarCoder2 15B is an open-weights language model from BigCode, released February 2024. Its published size is 15.96B parameters and a 16k-token context window. Licence: bigcode-openrail-m.

Catalogue identity notes

Maker repo bigcode/starcoder2-15b @46d4474: licence tag bigcode-openrail-m. 15,957,889,024 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-02: month the maker's Hugging Face repo was created (createdAt 2024-02-20). Base model; not StarCoder2-15B-Instruct. Ollama tags starcoder2:15b-q4_K_M, starcoder2:15b-q8_0 (sizes as listed on the tags page, observed 2026-09-29).

38.8±0.5#547 of 719 LLMsfrom 17 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of StarCoder2 15B, out of 828 with a score, then more from BigCode.

#modelscorewhereRAMreleased
637Starling-LM 7B alpha UC Berkeley (NEST)39±1local~6.1GB2023-11
638Granite 8B Code Base (4K) IBM39±1local~6.6GB2024-04
639StarCoder2 15B BigCode39±1local~11GB2024-02
640speechless-code-mistral-7b-v1.0 uukuguy39±1local~17GB2023-10
641Falcon-H1 3B Instruct TII39±4local~3.4GB2025-05
more from BigCode
521StarCoder2-15B-Instruct-v0.1 BigCode46±1local~11GB2024-04
698StarCoder2 7B BigCode35±1local~6.1GB2024-02
701StarCoder 15.5B BigCode35±1local~11GB2023-04

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #390 of 458 · 88% of its score · 9 tests

Beats DeepSeek R1 Distill Qwen 7B, Qwen3.5 2B, DeepSeek R1 Distill Llama 8B here, though they rank higher overall. Trails Qwen2.5 1.5B Instruct, WizardLM 13B V1.2, Falcon3 3B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval +246.3% ✎▁·▁▁▃▄▆▅█▃▁▁ beats 30/99easy · avg 79%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 47%sharp · at its level · several runs
HumanEval+37.8% ✎▁·▁▁▁▃▄▄▄█▃▁ beats 27/99easy · avg 74%▇▇▇▇▇▇▇ 18%sharp · at its level
MBPP +151.2% ✎█·▅▃▃▃▅▃▃▅·▅ beats 14/15easy · avg 82%▇▇ 5.6%at its level · several runs
HumanEvalSynthesize (HumanEvalPack, 6-language average)44.4% ✎▁·▂▁▂▆▅█▄▄▁▂ beats 26/35medium · avg 64%▇▇ 4.7%at its level
DS-1000 lab33.8% ✎▄▄·▄██▄█·█·▄ beats 10/12medium · avg 52%▇ 2.9%at its level
HumanEvalExplain (HumanEvalPack, 6-language average)17.0% ✎▄▇█▃▄▇▄▁▇▃·▁ beats 12/35medium · avg 49%▇ 2.8%typical
CRUXEval-I +147.4% ✎▃▅█▃▃▅·▃▃▅·▅ beats 14/15medium · avg 59%▇ 2.7%at its level · several runs
HumanEvalFix (HumanEvalPack, 6-language average)21.9% ✎▁·▁▁▄██▄█▁▅▅ beats 20/35medium · avg 56%▇ 2.6%sharp
1 more test
testresultvs the others in its rundifficultycountswhy
CRUXEval-O +146.7% ✎▃·▃██▅▃·▃▃▅▃ beats 15/15medium · avg 55%▇ 2.2%blunt · at its level · several runs
math▁▂▃▅▆▅▄█▄▁▁▁ #284 of 398 * · 1.9% of its score · 1 test

Beats Llama 3.1 Nemotron 70B Instruct, GPT-4 (0613), Athene-70B here, though they rank higher overall. Trails DeepSeek R1 Distill Qwen 1.5B, Phi-2 (2.7B) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K lab65.1% ✎▄▄▄·▄█·██·█▄ beats 12/12easy · avg 75%▇ 1.9%sharp · at its level
other tests2 not in a task · 10% of its score
testresultvs the others in its rundifficultycountswhy
EvoEval (overall)29.0% ✎▂▂▃▄▄█▁▄▅▃▁▃ beats 13/42medium · avg 55%▇▇▇ 6.9%at its level
BigCodeBench (Complete) lab38.3% ✎▂▁▁▄▂▅▇█▆▂▂▂ beats 22/58medium · avg 50%▇ 2.8%at its level

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run StarCoder2 15B · needs ~9.9 GB · from huggingface.co/second-state
    chat in the terminal
    llama cli -hf second-state/StarCoder2-15B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf second-state/StarCoder2-15B-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run StarCoder2 15B · needs ~9.9 GB · from huggingface.co/second-state

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/second-state/StarCoder2-15B-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run StarCoder2 15B · needs ~9.9 GB · from ollama.com
    chat in the terminal
    ollama run starcoder2:15b-q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run StarCoder2 15B · needs ~9 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/bigcode-starcoder2-15b-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/bigcode-starcoder2-15b-4bit

Official weights and documentation: huggingface.co/bigcode ↗

$ how the score was worked out · EvalPlus / EvalPerf published… 43%, Top Leaderboard Ranking = Top… 20%, StarCoder 2 and The Stack v2:… 17% and 2 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EvalPlus / EvalPerf published evaluation runs2▇▇▇▇▇▇▇▇▇▇▇▇▇ 43%Apache-2.0
Top Leaderboard Ranking = Top Coding Proficiency, Always? EvoEval: Evolving Coding Benchmarks via LLM — authors' own runs2▇▇▇▇▇▇ 20%CC BY 4.0
StarCoder 2 and The Stack v2: The Next Generation — authors' own runs · lab-reported6▇▇▇▇▇ 17%CC BY-SA 4.0
Granite Code Models: A Family of Open Foundation Models for Code Intelligence — authors' own runs6▇▇▇▇▇ 16%CC BY 4.0
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v4 · lab-reported1▇ 3%CC BY 4.0
prior · pull toward the middle▇ 0%
weighted average= 38.8

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json