👁 BigCode
BigCode
StarCoder2 15B
Current
10.3KDownloads673LikesFeb 2024Released16K tokensContextBigCode OpenRAIL-MLicense7 EntryQuality
StarCoder2 15B (15B parameters) requires approximately 13.8 GB of VRAM with Q5_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 16 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run StarCoder2 15B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "bigcode/starcoder2-15b" \
--hf-file "starcoder2-15b-Q5_K_M.gguf" \
-c 4096 -ngl 99Quick specs
Parameters15B
Architecturedense
Context16K tokens
Modalitycode
Min RAM5.9 GB
Rec. RAM10.8 GB (Q5_K_M)
LicenseBigCode OpenRAIL-M
FamilyStarCoder2
✓ Code
About this model
- •Model Summary
- •Limitations
- •Training
- •License
- •Citation
Related models
Your hardware
Detecting...
Quick picks
Best hardware
Top picks for StarCoder2 15B
Run this model
Quantization options
VRAM estimates by quant level
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.9 GB | Low | — |
Q3_K_S | 3 | 7.4 GB | Low | — |
NVFP4 | 4 | 8.4 GB | Medium | — |
Q4_K_M | 4 | 9.2 GB | Medium | — |
Q5_K_M | 5 | 10.8 GB | High | — |
Q6_K | 6 | 12.3 GB | High | — |
Q8_0 | 8 | 16.1 GB | Very High | — |
F16 | 16 | 30.7 GB | Maximum | — |
Quality benchmarks
StarCoder2 15B benchmark scores
Coding
SWE-bench Verified—
HumanEval+46.3%
Aider Polyglot—
LiveCodeBench—
Reasoning
MMLU-Pro15.0%
GPQA Diamond3.1%
MATH-5006.0%
ARC Challenge—
General
Chatbot Arena—
IFEval27.8%
Source: official · 2024-02-28
Hardware compatibility
Fit estimates across all hardware
Computing compatibility...
Memory breakdown
Reference: RTX 2060 6GB
Weights10.8 GB
KV Cache1.2 GB
Runtime1.2 GB
Headroom0.6 GB
Frequently asked questions
FAQ — StarCoder2 15B
See also