ethicalabs/BlossomTuneMLX-SmolLM2-135M-Instruct-Q8-NLP
By ethicalabs · ↓ 3 · ♥ 0 · Updated 2025-09-02
What's in this repo
This page lists the files in ethicalabs/BlossomTuneMLX-SmolLM2-135M-Instruct-Q8-NLP.
- No .onnx files in this repository
- 11 files, including 6 .json, 1 .md, 1 .jinja, 1 .txt, 1 .safetensors
- Last updated 2025-09-02
ONNX files
This repository has no .onnx files. See the full file list on the right for other formats.
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
ethicalabs/BlossomTuneMLX-SmolLM2-135M-Instruct-Q8-NLP
This model ethicalabs/BlossomTuneMLX-SmolLM2-135M-Instruct-Q8-NLP was converted to MLX format from HuggingFaceTB/SmolLM2-135M-Instruct-Q8-mlx using mlx-lm version 0.26.3.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("ethicalabs/BlossomTuneMLX-SmolLM2-135M-Instruct-Q8-NLP")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
Fine-tuning
This model has been fine-tuned with BlossomTuneLLM-MLX
INFO : aggregate_fit: received 10 results and 0 failures
INFO : Communication cost: 6.23 MB this round / 124.51 MB total
Server: Saving global adapter for round 10...
Fetching 8 files: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 8/8 [00:00<00:00, 8367.69it/s]
Global adapter and config saved to results/huggingfacetb-smollm2-135m-instruct-q8-mlx/server/2025-09-02_23-16-23/adapter_10
INFO : fit progress: (10, 0.0, {}, 239.02917212501052)
INFO : configure_evaluate: no clients selected, skipping evaluation
INFO :
INFO : [SUMMARY]
INFO : Run finished 10 round(s) in 239.03s
INFO : History (loss, centralized):
INFO : round 0: 0.0
INFO : round 1: 0.0
INFO : round 2: 0.0
INFO : round 3: 0.0
INFO : round 4: 0.0
INFO : round 5: 0.0
INFO : round 6: 0.0
INFO : round 7: 0.0
INFO : round 8: 0.0
INFO : round 9: 0.0
INFO : round 10: 0.0
INFO : History (metrics, distributed, fit):
INFO : {'train_loss': [(1, 2.2529776644706727),
INFO : (2, 1.6681898140907288),
INFO : (3, 1.5494979882240296),
INFO : (4, 1.4766268157958984),
INFO : (5, 1.4757164913415908),
INFO : (6, 1.387213920354843),
INFO : (7, 1.4945470476150513),
INFO : (8, 1.464623532295227),
INFO : (9, 1.4590632796287537),
INFO : (10, 1.4046799695491792)],
INFO : 'val_loss': [(1, 2.0296000242233276),
INFO : (2, 1.6557256400585174),
INFO : (3, 1.5062924563884734),
INFO : (4, 1.4948512375354768),
INFO : (5, 1.4645283639431),
INFO : (6, 1.4505432009696961),
INFO : (7, 1.4502118945121765),
INFO : (8, 1.4655221998691559),
INFO : (9, 1.4796700835227967),
INFO : (10, 1.429529356956482)]}
Model info
Created: 2025-09-02
Updated: 2025-09-02
View on Hugging Face