Instructions to use mrsandip/Alfa-Code with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mrsandip/Alfa-Code with Transformers:

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

pipe = pipeline("text-generation", model="mrsandip/Alfa-Code")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mrsandip/Alfa-Code", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use mrsandip/Alfa-Code 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 mrsandip/Alfa-Code:Q4_0
# Run inference directly in the terminal:
llama cli -hf mrsandip/Alfa-Code:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf mrsandip/Alfa-Code:Q4_0
# Run inference directly in the terminal:
llama cli -hf mrsandip/Alfa-Code:Q4_0

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf mrsandip/Alfa-Code:Q4_0
# Run inference directly in the terminal:
./build/bin/llama cli -hf mrsandip/Alfa-Code:Q4_0

Use Docker

docker model run hf.co/mrsandip/Alfa-Code:Q4_0

Llama new
LM Studio
Jan

vLLM

How to use mrsandip/Alfa-Code with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/mrsandip/Alfa-Code:Q4_0

SGLang

How to use mrsandip/Alfa-Code 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 "mrsandip/Alfa-Code" \
    --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": "mrsandip/Alfa-Code",
		"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 "mrsandip/Alfa-Code" \
        --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": "mrsandip/Alfa-Code",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use mrsandip/Alfa-Code with Ollama:
```
ollama run hf.co/mrsandip/Alfa-Code:Q4_0
```
Unsloth Desktop
Docker Model Runner
How to use mrsandip/Alfa-Code with Docker Model Runner:
```
docker model run hf.co/mrsandip/Alfa-Code:Q4_0
```

Lemonade

How to use mrsandip/Alfa-Code with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mrsandip/Alfa-Code:Q4_0

Run and chat with the model

lemonade run user.Alfa-Code-Q4_0

List all available models

lemonade list

Atomic Chat

mrsandip commited on 20 days ago

Commit

50f063a

verified ·

1 Parent(s): 07effa7

docs: readme v0.0.2 — gguf table

Browse files

Files changed (1) hide show

README.md +35 -6

README.md CHANGED Viewed

@@ -15,25 +15,54 @@ datasets:
 metrics:
 - code_eval
 - perplexity
-version: 0.4.0-trained
 ---
 # Alfa Code - Multimodal Coding GGUF
-RTX 3050 4GB: use tiny-50M demo. 1B+ needs cloud 8xH100.
 ## 1. Data
 `python scripts/download_data.py --dataset code_small_test --max-rows 500`
 Output: `data/processed/train.jsonl` {modality,prompt,think,answer}
 ## 2. Tokenizer + Train
-`python scripts/train_tokenizer.py` -> `tokenizers/alfa-32k.json` (verified 47KB)
 `pip install torch --index-url https://download.pytorch.org/whl/cu121`
 `python scripts/train.py` -> `outputs/tiny-50M`
-## 3. GGUF
-`git clone https://github.com/ggerganov/llama.cpp third_party/llama.cpp`
-`python scripts/convert_to_gguf.py --ckpt outputs/tiny-50M --quant Q4_K_M --out outputs/tiny-50M.gguf`
 ## 4. UI (reasoning live->collapsed + image/file/video)
 `python app/ui.py` -> http://127.0.0.1:7860

 metrics:
 - code_eval
 - perplexity
+version: 0.0.2
 ---
 # Alfa Code - Multimodal Coding GGUF
+Llama-based code models, trained on Kaggle T4. Ready-to-run GGUF weights below.
+## Download (ready weights)
+**GitHub release:** https://github.com/mrsandipmandal/alfa-code/releases/tag/v0.0.2
+**HuggingFace:** https://huggingface.co/mrsandip/Alfa-Code/tree/main/gguf
+| Model | Q8_0 (best) | Q4_0 (small) | Context |
+|---|---|---|---|
+| tiny-50M (56M) | tiny-50M-q8.gguf (60 MB) | tiny-50M-q4.gguf (32 MB) | 2048 |
+| base-1B (1.1B) | base-1B-q8.gguf (1.19 GB) | base-1B-q4.gguf (632 MB) | 2048 |
+| code-100k (1.39B) | ⏳ Kaggle upload pending | ⏳ | 8192 (YaRN → 131k) |
+```bash
+llama-server -m base-1B-q4.gguf -c 2048
+llama-server -m tiny-50M-q8.gguf -c 2048
+```
+Tokenizer (32k BPE): `tokenizers/alfa-32k.json` — also in the v0.0.2 release assets.
 ## 1. Data
 `python scripts/download_data.py --dataset code_small_test --max-rows 500`
 Output: `data/processed/train.jsonl` {modality,prompt,think,answer}
 ## 2. Tokenizer + Train
+`python scripts/train_tokenizer.py` -> `tokenizers/alfa-32k.json` (32000 vocab, ~419KB)
 `pip install torch --index-url https://download.pytorch.org/whl/cu121`
 `python scripts/train.py` -> `outputs/tiny-50M`
+Scale-up (Colab T4, ~2-4h): 5000+ rows, chat tuning, 3-5 epochs:
+`python scripts/download_data.py --dataset multimodal_mix --max-rows 5000 --overwrite`
+`python scripts/download_data.py --dataset chat_qa --max-rows 2000` (appends Q&A rows)
+`python scripts/train_hf.py --data data/processed/train.jsonl --out /content/outputs/tiny-50M --epochs 3 --config configs/tiny-50M.yaml --max-seq-len 2048 --batch-size 1`
+1B model (16GB+ GPU only, not RTX 3050 4GB): same commands with `--config configs/base-1B.yaml --out /content/outputs/base-1B` (1.12B params, ~2.2GB fp16).
+## 3. GGUF (no llama.cpp clone needed)
+`pip install gguf safetensors torch transformers`
+`python scripts/convert_to_gguf.py --ckpt outputs/tiny-50M --quant Q8_0` (Q8_0 ~55MB, Q4_0 ~32MB, F16 ~105MB)
+K-quants (Q4_K_M) need standard dims + `llama-quantize` binary — not offered (intermediate 1376).
+Run local (RTX 3050 4GB):
+`llama-server -m outputs/tiny-50M.gguf -c 2048` or Ollama Modelfile (`FROM ./outputs/tiny-50M.gguf`).
 ## 4. UI (reasoning live->collapsed + image/file/video)
 `python app/ui.py` -> http://127.0.0.1:7860