Mellum2 Instruct — GGUF (Q8_0)
This repository contains a GGUF Q8_0 quantization of
JetBrains/Mellum2-12B-A2.5B-Instruct, ready to run with
llama.cpp, Ollama, LM Studio, and
other GGUF-compatible runtimes.
This quantization (Q8_0): 8-bit. Effectively lossless — KL divergence from BF16 is ~0.016 and it keeps BF16's top token 95% of the time.
| File | Size |
|---|---|
Mellum2-12B-A2.5B-Instruct-Q8_0.gguf | 12.9 GB |
Mellum 2 Instruct is a Mixture-of-Experts assistant model (64 experts, 8 activated per token, 131,072-token context) that answers directly, without an externalized chain of thought. For the full model description, evaluation results, and architecture details, see the original model card: JetBrains/Mellum2-12B-A2.5B-Instruct.
Available quantizations
| Quantization | Description | Size | KLD vs BF16 ↓ | Top-token match ↑ |
|---|---|---|---|---|
BF16 | 16-bit, no quantization (reference) | 24.3 GB | — | — |
Q8_0 (this repo) | 8-bit, effectively lossless | 12.9 GB | 0.016 | 95.2% |
Q6_K | 6-bit k-quant, very high quality | 10.9 GB | 0.038 | 92.9% |
Q4_K_M | 4-bit k-quant, balanced (recommended) | 8.1 GB | 0.106 | 87.2% |
MXFP4_MOE | MXFP4 4-bit on MoE experts, smallest | 7.0 GB | 0.166 | 84.2% |
KL divergence and top-token agreement are measured against the BF16 logits on
Wikitext-2 (n_ctx=512); lower KLD / higher agreement means closer to the
unquantized model. (Perplexity is omitted here — it is unreliable for
instruction-tuned models on Wikitext-2, which is out of distribution.)
Download
hf download JetBrains/Mellum2-12B-A2.5B-Instruct-GGUF-Q8_0 Mellum2-12B-A2.5B-Instruct-Q8_0.gguf --local-dir .
Run with llama.cpp
# Pull and serve in one step (downloads the GGUF automatically)
llama-server -hf JetBrains/Mellum2-12B-A2.5B-Instruct-GGUF-Q8_0 \
--ctx-size 131072 \
--temp 0.6 --top-p 0.95 --top-k 20
# Or run a one-off prompt with a local file
llama-cli -m Mellum2-12B-A2.5B-Instruct-Q8_0.gguf \
--ctx-size 131072 \
--temp 0.6 --top-p 0.95 --top-k 20 \
-p "Write a Python function to reverse a string."
The server exposes an OpenAI-compatible API on http://localhost:8080/v1:
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8080/v1", api_key="llama.cpp")
chat_response = client.chat.completions.create(
model="JetBrains/Mellum2-12B-A2.5B-Instruct-GGUF-Q8_0",
messages=[
{"role": "user", "content": "Write a Python function to reverse a string."},
],
max_tokens=81920,
temperature=0.6,
top_p=0.95,
extra_body={"top_k": 20},
)
print(chat_response.choices[0].message.content)
Run with Ollama
ollama run hf.co/JetBrains/Mellum2-12B-A2.5B-Instruct-GGUF-Q8_0
License
Released under the Apache 2.0 license.
For the full model card, evaluation results, and architecture details, refer to the original model: JetBrains/Mellum2-12B-A2.5B-Instruct.