A coding-focused fine-tune of Qwen2.5-Coder-0.5B, trained via self-play reinforcement learning using 4 LLM judges and DeepSeek API evaluation. Trained on an AMD Radeon RX 9070 XT (ROCm 7.2) as part of the CloneOllama project.

Training Method: Self-Play

prompt → Student (Qwen2.5-Coder-0.5B) → generates code
         → Fast Python syntax scorer (1ms)
         → DeepSeek API reasoning judge (~3s)
         → Score ≥ 5.5 → LoRA rank 16 fine-tune

500 rounds, 397 accepted (79% acceptance rate)
Scores improved: 5.3 → 5.5 / 10 over training
Training data: Generated live by the student model, no human labeling
Prompts: Python coding, C# coding, algorithms, CS theory

Benchmarks (RX 9070 XT, Vulkan)

Metric	Value
Prompt speed	619 tokens/s
Generation speed	540 tokens/s
Model size	506 MB (Q8_0 GGUF) / 942 MB (FP16 safetensors)

Code Quality

Prompt	Output
`def is_palindrome(s):`	`s = s.lower().replace(" ", ""); return s == s[::-1]` ✅
`def fibonacci(n):`	`n = int(n); return [1, n]`
`def reverse_list(arr):`	`def reverse(arr, size): arr = [1,2,3,4,5]`
`def binary_search(arr, target):`	Partial structure

Judges Used During Training

Judge Model	Format	Gen Speed
DeepSeek-Coder-V2 16B	Q4_K_M GGUF	221 t/s
VibeThinker-3B	Q6_K GGUF	177 t/s
gemma4-coding	Q4_K_M GGUF	56 t/s
Qwen2.5-Coder-7B	i1-Q4_K_M GGUF	120 t/s

All judges ran concurrently as llama-server instances on different ports.

Files

File	Size	Format
`selfplay-coder-0.5B-Q8_0.gguf`	531 MB	GGUF Q8_0 (llama.cpp)
`model.safetensors`	942 MB	HuggingFace safetensors
`config.json`	—	Model config
`tokenizer.json`	11.4 MB	Qwen2.5 tokenizer

Usage

llama.cpp / Ollama

llama-server -m selfplay-coder-0.5B-Q8_0.gguf -ngl 99 --port 8080

Transformers (Python)

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Strangefrost/CloneOllama-selfplay-coder-0.5B")
tokenizer = AutoTokenizer.from_pretrained("Strangefrost/CloneOllama-selfplay-coder-0.5B")

prompt = "def fibonacci(n):"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0]))

Training Pipeline

Full pipeline scripts at model-kitchen/:

Script	Purpose
`selfplay_train.py`	Multi-judge self-play training
`prune.py`	Structured pruning + PEFT LoRA
`train_utils.py`	LoRA, distillation, calibration
`merge.py`	SLERP / TIES / DARE merging
`upcycle.py`	Dense → MoE conversion

Acknowledgments

Base model: Qwen/Qwen2.5-Coder-0.5B
Training platform: CloneOllama on AMD Radeon RX 9070 XT (ROCm 7.2)
Judge models: DeepSeek-Coder-V2, VibeThinker, gemma4-coding, Qwen2.5-Coder-7B