Holo-3.1-9B-Coder

Code-specialized adaptation of Hcompany/Holo-3.1-9B.

Evaluation

Model HumanEval+ pass@1 LiveCodeBench v2 pass@1
Holo-3.1-9B (base) 52.4% 31.5%
Holo-3.1-9B-Coder (this model) 65.2% (+12.8) 37.8% (+6.3)

LiveCodeBench v2 pass@1: 37.8% on the latest rented-GPU rerun (15 October 2026). Evaluation summary. Only the aggregate score was retained from that rerun.

Artifacts

Path Description
adapter/ LoRA adapter (r=8, alpha=16, q/v targets).
model.safetensors, config.json, tokenizer files Merged base + adapter model.
holo-9b-coder-Q4_K_M.gguf llama.cpp GGUF, Q4_K_M quantization.
holo-9b-coder-Q5_K_M.gguf llama.cpp GGUF, Q5_K_M quantization.
holo-9b-coder-Q6_K.gguf llama.cpp GGUF, Q6_K quantization.

Quantization details

Quantization was performed with llama.cpp after merging the adapter into the base model. The GGUF files use K-quant mixture schemes.

Loading the adapter

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = "Hcompany/Holo-3.1-9B"
model = AutoModelForCausalLM.from_pretrained(base, trust_remote_code=True, torch_dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(model, "josephmayo/Holo-3.1-9B-Coder", subfolder="adapter")
model = model.merge_and_unload()
tok = AutoTokenizer.from_pretrained(base, trust_remote_code=True)

Loading a GGUF with llama.cpp

./llama-cli -m holo-9b-coder-Q4_K_M.gguf -p "Return only executable Python code.\n\ndef factorial(n):" -n 256

Base model

  • Architecture: Qwen3.5-family text backbone (decoder-only transformer).
  • License: Apache-2.0.
  • Original model: Hcompany/Holo-3.1-9B

Limitations

  • Generated code should be reviewed before execution.
  • The model inherits the base model's knowledge cutoff and safety profile.
Downloads last month
561
Safetensors
Model size
9B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for josephmayo/Holo-3.1-9B-Coder

Quantized
(6)
this model
Quantizations
2 models