Instructions to use uukuguy/speechless-coder-ds-1.3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use uukuguy/speechless-coder-ds-1.3b with Transformers:

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

pipe = pipeline("text-generation", model="uukuguy/speechless-coder-ds-1.3b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("uukuguy/speechless-coder-ds-1.3b")
model = AutoModelForCausalLM.from_pretrained("uukuguy/speechless-coder-ds-1.3b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use uukuguy/speechless-coder-ds-1.3b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "uukuguy/speechless-coder-ds-1.3b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "uukuguy/speechless-coder-ds-1.3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/uukuguy/speechless-coder-ds-1.3b

SGLang

How to use uukuguy/speechless-coder-ds-1.3b 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 "uukuguy/speechless-coder-ds-1.3b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "uukuguy/speechless-coder-ds-1.3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "uukuguy/speechless-coder-ds-1.3b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "uukuguy/speechless-coder-ds-1.3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use uukuguy/speechless-coder-ds-1.3b with Docker Model Runner:
```
docker model run hf.co/uukuguy/speechless-coder-ds-1.3b
```

speechless-coder-ds-1.3b

Use the following dataset to fine-tune deepseek-ai/deepseek-coder-1.3b in order to improve the model's reasoning and planning abilities.

context window length: 8192 max_tokens > 128 && < 8192

Total 185,193 samples 426 MB

ise-uiuc/Magicoder-OSS-Instruct-75K 75,186 samples
ise-uiuc/Magicoder-Evol-Instruct-110K 110,007 samples

50 samples/T=0.2/MaxTokens=512/Top_P=0.95

Code: https://github.com/uukuguy/speechless

How to Prompt the Model

This model accepts the Alpaca instruction format.

For example:

You are an intelligent programming assistant.

### Instruction:
Implement a linked list in C++

### Response:

HumanEval

Metric	Value
humaneval-python

Big Code Models Leaderboard

CodeLlama-34B-Python: 53.29

CodeLlama-34B-Instruct: 50.79

CodeLlama-13B-Instruct: 50.6

CodeLlama-34B: 45.11

CodeLlama-13B-Python: 42.89

CodeLlama-13B: 35.07

BigCode Eval

0.205055

metrics_humanevalfixtests-cpp: "pass@1": 0.054878048780487805
metrics_humanevalfixtests-go: "pass@1": 0.054878048780487805
metrics_humanevalfixtests-java: "pass@1": 0.042682926829268296
metrics_humanevalfixtests-js: "pass@1": 0.0975609756097561
metrics_humanevalfixtests-python: "pass@1": 0.06707317073170732
metrics_humanevalfixtests-rust: "pass@1": 0.018292682926829267

0.332906

metrics_humanevalsynthesize-cpp: "pass@1": 0.3475609756097561
metrics_humanevalsynthesize-go: "pass@1": 0.25609756097560976
metrics_humanevalsynthesize-java: "pass@1": 0.3353658536585366
metrics_humanevalsynthesize-js: "pass@1": 0.35365853658536583
metrics_humanevalsynthesize-python: "pass@1": 0.4024390243902439
metrics_humanevalsynthesize-rust: "pass@1": 0.20121951219512196
metrics_mbpp: "pass@1": 0.434

LMEval

Open LLM Leaderboard

Metric	Value
ARC
HellaSwag
MMLU
TruthfulQA
Average

Downloads last month: 888

Datasets used to train uukuguy/speechless-coder-ds-1.3b

Collection including uukuguy/speechless-coder-ds-1.3b

DeepSeek-based Models

Collection

3 items • Updated Jan 15, 2024

Evaluation results

pass@1 on HumanEval
self-reported