Instructions to use CubicLabs/AXL-Chat-10M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CubicLabs/AXL-Chat-10M with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="CubicLabs/AXL-Chat-10M")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("CubicLabs/AXL-Chat-10M", dtype="auto")

llama-cpp-python

How to use CubicLabs/AXL-Chat-10M with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="CubicLabs/AXL-Chat-10M",
	filename="axl-chat-f16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use CubicLabs/AXL-Chat-10M with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf CubicLabs/AXL-Chat-10M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf CubicLabs/AXL-Chat-10M:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf CubicLabs/AXL-Chat-10M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf CubicLabs/AXL-Chat-10M:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf CubicLabs/AXL-Chat-10M:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf CubicLabs/AXL-Chat-10M:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf CubicLabs/AXL-Chat-10M:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf CubicLabs/AXL-Chat-10M:Q4_K_M

Use Docker

docker model run hf.co/CubicLabs/AXL-Chat-10M:Q4_K_M

LM Studio
Jan

vLLM

How to use CubicLabs/AXL-Chat-10M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CubicLabs/AXL-Chat-10M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CubicLabs/AXL-Chat-10M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CubicLabs/AXL-Chat-10M:Q4_K_M

SGLang

How to use CubicLabs/AXL-Chat-10M with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "CubicLabs/AXL-Chat-10M" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CubicLabs/AXL-Chat-10M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "CubicLabs/AXL-Chat-10M" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CubicLabs/AXL-Chat-10M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use CubicLabs/AXL-Chat-10M with Ollama:
```
ollama run hf.co/CubicLabs/AXL-Chat-10M:Q4_K_M
```

Unsloth Studio

How to use CubicLabs/AXL-Chat-10M with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for CubicLabs/AXL-Chat-10M to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for CubicLabs/AXL-Chat-10M to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for CubicLabs/AXL-Chat-10M to start chatting

Atomic Chat new
Docker Model Runner
How to use CubicLabs/AXL-Chat-10M with Docker Model Runner:
```
docker model run hf.co/CubicLabs/AXL-Chat-10M:Q4_K_M
```

Lemonade

How to use CubicLabs/AXL-Chat-10M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull CubicLabs/AXL-Chat-10M:Q4_K_M

Run and chat with the model

lemonade run user.AXL-Chat-10M-Q4_K_M

List all available models

lemonade list

Restrictated commited on 3 days ago

Commit

2ffcd78

verified ·

1 Parent(s): 8299d9f

Update README.md

Browse files

Files changed (1) hide show

README.md +18 -53

README.md CHANGED Viewed

@@ -6,7 +6,7 @@ tags:
   - code-generation
   - multi-scale-transformer
   - cpu-optimized
-  - koinic
   - pytorch
   - llama
   - gguf
@@ -15,7 +15,7 @@ tags:
 pipeline_tag: text-generation
 library_name: transformers
 datasets:
-  - koinic/axl-chat-pairs
 widget:
   - text: "User: How do I read a CSV in Python?\nAssistant:"
 model-index:
@@ -31,13 +31,13 @@ model-index:
 # AXL-Chat-10M
-Conversational AI. 9.9M params. PPL 1.02. Context 512 bytes. 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 | 10M |
 | Optimizer | Lion |
@@ -52,8 +52,8 @@ Conversational AI. 9.9M params. PPL 1.02. Context 512 bytes. Part of the AXL mod
 ### Sources
-- **Repository:** [GitHub](https://github.com/Koinic/AXL)
-- **Organization:** [KoinicLabs](https://huggingface.co/KoinicLabs)
 ## Uses
@@ -61,7 +61,7 @@ Conversational AI. 9.9M params. PPL 1.02. Context 512 bytes. Part of the AXL mod
 Conversational AI for programming Q&A.
-```python
 import torch
 from multiscale_transformer.model.model import MultiScaleTransformer
 from multiscale_transformer.training.tokenizer import ByteTokenizer
@@ -74,7 +74,6 @@ 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
@@ -100,14 +99,6 @@ Retrained with Lion on 10MB chat pairs. 216 steps in 10 min. Covers code Q&A, ge
 Byte-level tokenization with vocabulary size 258 (256 bytes + BOS + EOS). No vocabulary training required.
-### Speeds, Sizes, Times
-| Metric | Value |
-|--------|-------|
-| Training Steps | 216 |
-| Training Time | 10 min |
-| Final Loss | 0.3650 |
 ## Evaluation
 ### Metrics
@@ -116,61 +107,36 @@ Perplexity on held-out Python code using byte-level tokenization.
 ### Results
-| Metric | Value |
-|--------|-------|
-| Perplexity (byte-level) | 1.02 |
-| Final Loss | 0.3650 |
-| Training Steps | 216 |
-| Training Time | 10 min |
-**Summary:** Good for code explanation and Q&A.
 ## Environmental Impact
-| Property | Value |
-|----------|-------|
-| Hardware | AMD Ryzen 5 5600G |
-| Hours Used | 0.167 |
-| Carbon Emitted | 0.0070 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-Chat-10M - 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-chat-10m -f Modelfile
 ollama run axl-chat-10m "def fibonacci():"
-```
 ### With Python
-```python
 import torch
 from multiscale_transformer.model.config import load_config
 from multiscale_transformer.model.model import MultiScaleTransformer
@@ -185,5 +151,4 @@ 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()))
-```

   - code-generation
   - multi-scale-transformer
   - cpu-optimized
+  - cubic
   - pytorch
   - llama
   - gguf
 pipeline_tag: text-generation
 library_name: transformers
 datasets:
+  - cubic/axl-chat-pairs
 widget:
   - text: "User: How do I read a CSV in Python?\nAssistant:"
 model-index:
 # AXL-Chat-10M
+Conversational AI. 9.9M params. PPL 1.02. Context 512 bytes. Part of the AXL model family by [CubicLabs](https://huggingface.co/CubicLabs).
 ## Model Details
 | Property | Value |
 |----------|-------|
+| Developed by | [CubicLabs](https://huggingface.co/CubicLabs) |
 | Architecture | Multi-Scale Transformer |
 | Parameters | 10M |
 | Optimizer | Lion |
 ### Sources
+- **Repository:** https://github.com/Cubic/AXL
+- **Organization:** https://huggingface.co/CubicLabs
 ## Uses
 Conversational AI for programming Q&A.
+Example Usage:
 import torch
 from multiscale_transformer.model.model import MultiScaleTransformer
 from multiscale_transformer.training.tokenizer import ByteTokenizer
 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
 Byte-level tokenization with vocabulary size 258 (256 bytes + BOS + EOS). No vocabulary training required.
 ## Evaluation
 ### Metrics
 ### Results
+Perplexity (byte-level): 1.02
+Final Loss: 0.3650
+Training Steps: 216
+Training Time: 10 min
 ## Environmental Impact
+Hardware: AMD Ryzen 5 5600G
+Hours Used: 0.167
+Carbon Emitted: 0.0070 kg CO2
+Cloud Provider: None (local CPU)
 ## Citation
 @misc{axl_2026,
   title={AXL: AXL-Chat-10M - Multi-Scale Transformer for CPU Code Generation},
+  author={Cubic},
   year={2026},
+  url={https://huggingface.co/CubicLabs}
 }
 ## How to Get Started
 ### With Ollama
 ollama create axl-chat-10m -f Modelfile
 ollama run axl-chat-10m "def fibonacci():"
 ### With Python
 import torch
 from multiscale_transformer.model.config import load_config
 from multiscale_transformer.model.model import MultiScaleTransformer
 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()))