StarCoder2

What StarCoder2 is for

StarCoder2 is trained for code. BigCode released it in 2024-02 with 3 sizes (3B, 7B, 15B), 16K of context and a BigCode Open RAIL-M licence. Context length matters more here than for chat, because a coding assistant that cannot see the whole file is guessing. The catalogue records its strength as code completion (VS Code, JetBrains), and it is the class of model an editor plugin such as Continue or Tabby points at.

Sizes and memory

The 3B needs about 3 GB of weights at Q4_K_M and the 15B about 10 GB, a spread of 7 GB across the family. The catalogue's default pick is 7B. Recommended VRAM adds roughly half again on top of the weights for the KV cache and the runtime.

Size	Weights at Q4_K_M	Recommended VRAM	Smallest card in our guide that holds it
3B	3 GB	5 GB	NVIDIA Jetson Orin Nano 8GB
7B	5 GB	8 GB	NVIDIA Jetson Orin Nano 8GB
15B	10 GB	15 GB	NVIDIA RTX 3060 12GB

Hardware that runs StarCoder2

Every one of the 19 cards with dedicated memory in our hardware guide holds at least one size of StarCoder2 at Q4_K_M. The cheapest that does is the NVIDIA Jetson Orin Nano 8GB at about $199, running the 7B.

The largest size any of them holds is the 15B at 10 GB, on the NVIDIA RTX 3060 12GB (12 GB at 360 GB/s). Bandwidth, not capacity, sets the speed once a size fits, so a card further down the list with faster memory will generate more tokens per second on the same weights than a larger, slower one.

What people use it for

The catalogue lists 3 recorded uses for StarCoder2: code completion, fill-in-middle, VS Code extensions. It is tagged code, completion, fill-in-middle.

Where StarCoder2 sits in the StarCoder family

StarCoder2 is the only StarCoder entry in our catalogue, so there is no in-family alternative to weigh it against; the comparison to make is against the models listed at the foot of this page.

It ships under the BigCode Open RAIL-M, which is neither a standard open-source licence nor a closed one. Read the terms before shipping anything commercial on top of it.

Context window: 16K

16K tokens comes to roughly 12,000 words at the usual three-quarters-of-a-word-per-token ratio: a long chat, a single source file, one chapter. It is the band where a retrieval step stops being optional, because the interesting documents do not fit. The recommended VRAM figure above assumes a working context rather than the maximum, and filling this one will push past it.

How to run StarCoder2 locally

Our catalogue does not record registry tags, so look the current tag up in the Ollama library or on Hugging Face before pasting these. The commands below are the shape of the workflow, not a copy-and-paste recipe.

Option 1: Ollama (simplest)

# find the tag for StarCoder2 at ollama.com/library
ollama run <tag>

Option 2: Mullama (production)

mullama pull <tag>
mullama run <tag>

Option 3: llama.cpp (CLI)

# download a GGUF from Hugging Face, searching for "StarCoder2 GGUF"
./llama-cli -m model.Q4_K_M.gguf -p "Hello, AI!"

Option 4: Python with Mullama or llama-cpp-python

from mullama import Model, Context
model = Model.load("model.Q4_K_M.gguf", n_gpu_layers=99)
ctx = Context(model, n_ctx=4096)
print(ctx.generate("Hello, AI!", 256))

Sources

Model record from src/data/models.json (2026-06-29); card capacities from src/data/gpus.json (2026-06-29). Upstream: Curated from ollama.com/library + community benchmarks + paperswithcode + huggingface.