Skip to content
localmodel.run

Text model · Qwen2.5-Coder

Qwen2.5 Coder 14B requirements

Qwen2.5-Coder family · 14B params · released Nov 2024 · 16.8M Ollama pulls. Minimum to run at Q4_K_M: Nvidia GeForce RTX 3060 (12GB).

LicenseApache-2.0· Commercial OK↓ 2.3M/mo♥ 163on HuggingFace
Q4_K_M
8.37 GB
Q8_0
14.62 GB
Total @ Q4 (4k)
~10.1 GB
Context
32 k

Quantization sizes

GGUF quantson disk
QuantizationSize on disk
Q2_K5.9 GB est
Q3_K_M6.8 GB est
Q4_K_M (default)8.37 GB
Q5_K_M10 GB est
Q6_K11.5 GB est
Q8_014.62 GB
FP1628 GB est

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run qwen2.5-coder:14b
llama.cpp
$ llama-cli -hf bartowski/Qwen2.5-Coder-14B-Instruct-GGUF:Q4_K_M
LM Studio
$ lms get bartowski/Qwen2.5-Coder-14B-Instruct-GGUF

Which devices can run Qwen2.5 Coder 14B?

FAQ

How much VRAM or RAM does Qwen2.5 Coder 14B need?

At Q4_K_M, Qwen2.5 Coder 14B needs about 10.1 GB (weights ~8.37 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~16.3 GB.

Can Qwen2.5 Coder 14B run on a laptop?

Qwen2.5 Coder 14B is large; you need a 24 GB+ GPU or a 32-48 GB Mac at Q4_K_M.

Can I use Qwen2.5 Coder 14B commercially?

Yes. Qwen2.5 Coder 14B is licensed Apache-2.0, which permits commercial use.

Qwen2.5-Coder, code-specialized. Q4_K_M and Q8_0 sizes from the bartowski GGUF repo. 32K native context, 128K with YaRN.

Sources

Memory figures are estimates. See methodology.