Instructions to use josephmayo/ZAYA1-8B-Coder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use josephmayo/ZAYA1-8B-Coder with Transformers:

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

pipe = pipeline("text-generation", model="josephmayo/ZAYA1-8B-Coder")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("josephmayo/ZAYA1-8B-Coder")
model = AutoModelForCausalLM.from_pretrained("josephmayo/ZAYA1-8B-Coder", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use josephmayo/ZAYA1-8B-Coder with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "josephmayo/ZAYA1-8B-Coder"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "josephmayo/ZAYA1-8B-Coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/josephmayo/ZAYA1-8B-Coder

SGLang

How to use josephmayo/ZAYA1-8B-Coder 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 "josephmayo/ZAYA1-8B-Coder" \
    --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": "josephmayo/ZAYA1-8B-Coder",
		"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 "josephmayo/ZAYA1-8B-Coder" \
        --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": "josephmayo/ZAYA1-8B-Coder",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use josephmayo/ZAYA1-8B-Coder with Docker Model Runner:
```
docker model run hf.co/josephmayo/ZAYA1-8B-Coder
```

josephmayo commited on May 27

Commit

0a6bb6c

verified ·

1 Parent(s): 7ca7233

Update model card with technical eval details

Browse files

Files changed (1) hide show

README.md +51 -5

README.md CHANGED Viewed

@@ -4,13 +4,13 @@ library_name: transformers
 tags:
 - zaya
 - lora
-- coding
 - merged
 ---
-# ZAYA1-8B Coding Standalone
-Merged standalone model from `Zyphra/ZAYA1-8B` and `josephmayo/ZAYA1-8B-Coding-LoRA`.
 ## Evaluation Gate
@@ -18,11 +18,47 @@ The adapter was evaluated against the base model on 50 Python coding prompts wit
 - Base average: 2.36 / 10
 - LoRA average: 4.76 / 10
 - Improved prompts: 39 / 50
-- Full-scale improvement: 24.00%
 - Merge threshold: 20.00%
 - Merge decision: true
 Evaluation artifacts are included under `eval/`:
 - `eval/eval_summary.json`
@@ -30,4 +66,14 @@ Evaluation artifacts are included under `eval/`:
 - `eval/base_outputs.jsonl`
 - `eval/lora_outputs.jsonl`
-Note: ZAYA uses a custom `model_type = zaya`; loading the original architecture requires Zyphra's Transformers implementation. A patched `zaya_patched_config.json` is also included for the requested compatibility artifact.

 tags:
 - zaya
 - lora
+- coder
 - merged
 ---
+# ZAYA1-8B Coder
+Merged Coder model from `Zyphra/ZAYA1-8B` and `josephmayo/ZAYA1-8B-Coder-LoRA`. This repo contains the adapter merged into the base weights as normal safetensors shards.
 ## Evaluation Gate
 - Base average: 2.36 / 10
 - LoRA average: 4.76 / 10
+- Absolute score delta: +2.40 / 10
+- Full-scale lift: 24.00%
+- Relative lift over base average: 101.69%
 - Improved prompts: 39 / 50
 - Merge threshold: 20.00%
 - Merge decision: true
+Full-scale lift is the required notebook metric:
+```text
+((lora_avg - base_avg) / 10) * 100
+((4.76 - 2.36) / 10) * 100 = 24.00%
+```
+## Scoring Heuristic
+Each response was scored out of 10:
+- `def` present: 2 points
+- `class` present: 1 point
+- `return` present: 1 point
+- `import` or `from` present: 1 point
+- fenced code block present: 1 point
+- output length greater than 100 characters: 1 point
+- Python AST parse validity: 3 points
+## Architecture Notes
+ZAYA uses a custom `model_type = zaya`; it is not weight-compatible with `LlamaForCausalLM` despite similar naming in some configs. During evaluation and merge, the real ZAYA architecture was loaded using Zyphra's Transformers implementation:
+```bash
+pip install git+https://github.com/Zyphra/transformers.git@zaya1
+```
+The LoRA adapter contains 160 tensors targeting:
+- `self_attn.o_proj`
+- `zaya_block.router.down_proj`
+The merge was performed after the evaluation gate passed, then the merged model was saved to safetensors shards with tokenizer and generation config.
 Evaluation artifacts are included under `eval/`:
 - `eval/eval_summary.json`
 - `eval/base_outputs.jsonl`
 - `eval/lora_outputs.jsonl`
+## Included Files
+- `model-00001-of-00005.safetensors` through `model-00005-of-00005.safetensors`
+- `model.safetensors.index.json`
+- `config.json`
+- `generation_config.json`
+- tokenizer files
+- `zaya_patched_config.json`
+- evaluation outputs under `eval/`
+The GGUF quantized release is available at `josephmayo/ZAYA1-8B-Coder-GGUF`.