---
license: apache-2.0
language:
  - code
tags:
  - code-generation
  - multi-scale-transformer
  - cpu-optimized
  - koinic
  - pytorch
  - llama
  - gguf
  - byte-level
pipeline_tag: text-generation
library_name: transformers
datasets:
  - bigcode/starcoderdata
  - theblackcat102/evol-codealpaca-v1
widget:
  - text: "# Find GCD of two numbers\ndef gcd(a, b):"
model-index:
  - name: AXL-Reasoning-70M
    results:
      - task:
          type: text-generation
        metrics:
          - name: Perplexity (byte-level)
            type: perplexity
            value: 1.93
---

# AXL-Reasoning-70M

CoT reasoning (SGD). 70M params. PPL 1.93 Part of the AXL model family by [KoinicLabs](https://huggingface.co/KoinicLabs).

## Model Details

| Property | Value |
|----------|-------|
| Developed by | [KoinicLabs](https://huggingface.co/KoinicLabs) |
| Architecture | Multi-Scale Transformer |
| Parameters | 70M |
| Optimizer | SGD |
| Attention | SDPA |
| Vocab Size | 258 (byte-level) |
| Context Window | 512 bytes |
| d_model | 512 |
| Attention Heads | 4 |
| Layers per Scale | 5 |
| Downsample Factors | [1, 2, 4] |
| License | Apache 2.0 |

### Sources

- **Repository:** [GitHub](https://github.com/Koinic/AXL)
- **Organization:** [KoinicLabs](https://huggingface.co/KoinicLabs)

## Uses

### Direct Use

Chain-of-thought reasoning (SGD baseline).

```python
import torch
from multiscale_transformer.model.model import MultiScaleTransformer
from multiscale_transformer.training.tokenizer import ByteTokenizer
ckpt = torch.load("axl_reasoning_70m.pt", map_location="cpu")
model = MultiScaleTransformer(config)
model.load_state_dict(ckpt["model_state_dict"])
model.eval()
tokenizer = ByteTokenizer()
ids = torch.tensor([tokenizer.encode("def hello():")], dtype=torch.long)
with torch.no_grad():
    out = model.generate(ids, max_new_tokens=50, temperature=0.8)
print(tokenizer.decode(out[0].tolist()))
```

### Out-of-Scope Use

Not for production code generation. Use the Lion version for better results. For integration with tools like Continue.dev, LlamaIndex, or LangChain, use the Python API server which provides OpenAI-compatible endpoints.

## Bias, Risks, and Limitations

Byte-level perplexity is not comparable to BPE-level perplexity. SGD baseline. Use AXL-Reasoning-Lion for better results. Note: GGUF files for Ollama use a simplified single-stack encoder. For full AXL quality, use the Python API server.

### Recommendations

- Use for prototyping and experimentation, not production code generation.
- Byte-level perplexity (258 vocab) is not comparable to BPE-level perplexity (32K vocab).
- For better results, use the Lion-optimized version if available.

## Training Details

### Training Data

SGD 60 min. 1478 steps. Lion version achieves PPL 1.93 in 20 min.

### Preprocessing

Byte-level tokenization with vocabulary size 258 (256 bytes + BOS + EOS). No vocabulary training required.

### Speeds, Sizes, Times

| Metric | Value |
|--------|-------|
| Training Steps | 1478 |
| Training Time | 60 min |
| Final Loss | 0.6384 |

## Evaluation

### Metrics

Perplexity on held-out Python code using byte-level tokenization.

### Results

| Metric | Value |
|--------|-------|
| Perplexity (byte-level) | 1.93 |
| Final Loss | 0.6384 |
| Training Steps | 1478 |
| Training Time | 60 min |

**Summary:** SGD reasoning baseline. Lion version is significantly better.

## Environmental Impact

| Property | Value |
|----------|-------|
| Hardware | AMD Ryzen 5 5600G |
| Hours Used | 1.001 |
| Carbon Emitted | 0.0420 kg CO2 |
| Cloud Provider | None (local CPU) |

## Technical Specifications

### Model Architecture

Multi-Scale Transformer with three parallel encoder stacks at resolution scales 1x, 2x, and 4x. Cross-scale attention connects all scale pairs. Adaptive gating fusion. SwiGLU feed-forward. RoPE positional encoding.

### Compute Infrastructure

| Property | Value |
|----------|-------|
| Hardware | AMD Ryzen 5 5600G (6 cores, 12 threads) |
| RAM | 16 GB |
| GPU | None (CPU-only) |

## Citation

```bibtex
@misc{axl_2026,
  title={AXL: AXL-Reasoning-70M - Multi-Scale Transformer for CPU Code Generation},
  author={Koinic},
  year={2026},
  url={https://huggingface.co/KoinicLabs}
}
```

## How to Get Started

### With Ollama

```bash
ollama create axl-reasoning-70m -f Modelfile
ollama run axl-reasoning-70m "def fibonacci():"
```

### With Python

```python
import torch
from multiscale_transformer.model.config import load_config
from multiscale_transformer.model.model import MultiScaleTransformer
from multiscale_transformer.training.tokenizer import ByteTokenizer
config = load_config("config.json")
model = MultiScaleTransformer(config)
ckpt = torch.load("axl_reasoning_70m.pt", map_location="cpu")
model.load_state_dict(ckpt["model_state_dict"])
model.eval()
tokenizer = ByteTokenizer()
prompt = "def fibonacci():"
ids = torch.tensor([tokenizer.encode(prompt)], dtype=torch.long)
with torch.no_grad():
    out = model.generate(ids, max_new_tokens=100, temperature=0.8, top_k=40)
print(tokenizer.decode(out[0].tolist()))
```