150 lines
4.6 KiB
Markdown
150 lines
4.6 KiB
Markdown
---
|
||
license: apache-2.0
|
||
language:
|
||
- pl
|
||
- en
|
||
- multilingual
|
||
base_model:
|
||
- speakleash/Bielik-11B-v3.0-Instruct
|
||
library_name: mlx
|
||
pipeline_tag: text-generation
|
||
tags:
|
||
- mlx
|
||
- apple-silicon
|
||
- bielik
|
||
- polish
|
||
- bfloat16
|
||
- quantized
|
||
- bf16/fp16
|
||
inference: false
|
||
widget:
|
||
- text: Wyjaśnij krótko różnicę między diagnostyką różnicową a rozpoznaniem.
|
||
example_title: Polish instruction prompt
|
||
- text: Podsumuj najważniejsze ryzyka w planie wdrożenia.
|
||
example_title: Polish reasoning prompt
|
||
---
|
||
|
||
# Bielik-11B-v3.0-mlx-bf16
|
||
|
||
`Bielik-11B-v3.0-mlx-bf16` is an MLX BF16/FP16 packaging of `speakleash/Bielik-11B-v3.0-Instruct` for Polish and multilingual instruction-style generation on Apple Silicon.
|
||
|
||
## Intended use
|
||
|
||
- Local text generation and chat-style prompting on Apple Silicon
|
||
- MLX-LM experimentation with the declared upstream model family
|
||
- Offline or operator-controlled inference workflows
|
||
|
||
## Out of scope
|
||
|
||
- Safety-critical decisions without domain expert review
|
||
- Claims of benchmark superiority not backed by published evaluation data
|
||
- Non-MLX runtime guarantees; this card documents the shipped HF checkpoint, not every possible serving stack
|
||
|
||
## Training and conversion metadata
|
||
|
||
| Parameter | Value |
|
||
|---|---|
|
||
| Repository | `LibraxisAI/Bielik-11B-v3.0-mlx-bf16` |
|
||
| Base model | `speakleash/Bielik-11B-v3.0-Instruct` |
|
||
| Task | `text-generation` |
|
||
| Library | `mlx` |
|
||
| Format | MLX / Apple Silicon checkpoint |
|
||
| Quantization | BF16/FP16 |
|
||
| Architecture | LlamaForCausalLM |
|
||
| Model files | 5 |
|
||
| Config model_type | `llama` |
|
||
|
||
This card only reports metadata present in the Hugging Face repository, existing card frontmatter, or public config files. Missing benchmark, dataset, or training-run details are left explicit rather than reconstructed.
|
||
|
||
## Usage
|
||
|
||
### CLI
|
||
|
||
```bash
|
||
pip install mlx-lm
|
||
|
||
mlx_lm.generate \
|
||
--model LibraxisAI/Bielik-11B-v3.0-mlx-bf16 \
|
||
--prompt "Opisz krótko objawy odwodnienia u psa i kiedy pilnie skontaktować się z lekarzem weterynarii." \
|
||
--max-tokens 400
|
||
```
|
||
|
||
### Python
|
||
|
||
```python
|
||
from mlx_lm import load, generate
|
||
|
||
model, tokenizer = load("LibraxisAI/Bielik-11B-v3.0-mlx-bf16")
|
||
|
||
prompt = "Opisz krótko objawy odwodnienia u psa i kiedy pilnie skontaktować się z lekarzem weterynarii."
|
||
response = generate(model, tokenizer, prompt=prompt, max_tokens=400)
|
||
print(response)
|
||
```
|
||
|
||
### Multi-turn with the chat template
|
||
|
||
This checkpoint follows the tokenizer/chat-template contract inherited from `speakleash/Bielik-11B-v3.0-Instruct` when the
|
||
template is present in the repository:
|
||
|
||
```python
|
||
from mlx_lm import load, generate
|
||
|
||
model, tokenizer = load("LibraxisAI/Bielik-11B-v3.0-mlx-bf16")
|
||
|
||
messages = [
|
||
{"role": "user", "content": "Opisz krótko objawy odwodnienia u psa i kiedy pilnie skontaktować się z lekarzem weterynarii."},
|
||
]
|
||
prompt = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
|
||
response = generate(model, tokenizer, prompt=prompt, max_tokens=400)
|
||
print(response)
|
||
```
|
||
|
||
## Example output
|
||
|
||
No public sample output is currently declared for this checkpoint. Run the usage example above against your own prompt or audio/image input to inspect behavior.
|
||
|
||
## Quantization notes
|
||
|
||
| Aspect | Original/base checkpoint | This checkpoint |
|
||
|---|---|---|
|
||
| Lineage | `speakleash/Bielik-11B-v3.0-Instruct` | `LibraxisAI/Bielik-11B-v3.0-mlx-bf16` |
|
||
| Runtime target | Upstream runtime format | MLX on Apple Silicon |
|
||
| Quantization | Base precision or upstream-declared format | BF16/FP16 |
|
||
| Published quality delta | Not declared in public metadata | Not declared in public metadata |
|
||
|
||
## Limitations
|
||
|
||
- No public benchmarks for this checkpoint are declared in the model metadata.
|
||
- No public benchmark claims are made by this card unless listed in the frontmatter.
|
||
- Validate outputs on your own domain data before relying on this checkpoint.
|
||
- Memory use and speed depend heavily on the exact Apple Silicon generation, unified-memory size, and prompt length.
|
||
|
||
## License
|
||
|
||
`apache-2.0`. Check the upstream/base model license as well when a base model is declared.
|
||
|
||
## Citation
|
||
|
||
```bibtex
|
||
@misc{libraxisai-bielik-11b-v3-0-mlx-bf16,
|
||
title = {Bielik-11B-v3.0-mlx-bf16},
|
||
author = {LibraxisAI},
|
||
year = {2026},
|
||
howpublished = {\url{https://huggingface.co/LibraxisAI/Bielik-11B-v3.0-mlx-bf16}},
|
||
note = {MLX checkpoint published by LibraxisAI}
|
||
}
|
||
```
|
||
|
||
## Inference tested on
|
||
|
||
[`LibraxisAI/mlx-batch-server`](https://github.com/LibraxisAI/mlx-batch-server)
|
||
|
||
## Related
|
||
|
||
- Base model: [`speakleash/Bielik-11B-v3.0-Instruct`](https://huggingface.co/speakleash/Bielik-11B-v3.0-Instruct)
|
||
|
||
---
|
||
|
||
𝚅𝚒𝚋𝚎𝚌𝚛𝚊𝚏𝚝𝚎𝚍. with AI Agents by VetCoders (c)2024-2026 LibraxisAI
|
||
|