Text Generation
MLX
Safetensors
English
pretraining
from-scratch
small-language-model
post-training
silicon
Instructions to use OpenSML/OpenSML-150M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use OpenSML/OpenSML-150M with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # if on a CUDA device, also pip install mlx[cuda] # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("OpenSML/OpenSML-150M") prompt = "Once upon a time in" text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- MLX LM
How to use OpenSML/OpenSML-150M with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "OpenSML/OpenSML-150M" --prompt "Once upon a time"
- Atomic Chat
Download evaluations/post_training_20260927.json from OpenSML/OpenSML-150M: direct link, hf CLI and curl.
- Browser
- Download file 6.48 kB
-
https://huggingface.co/OpenSML/OpenSML-150M/resolve/main/evaluations/post_training_20260927.json
- Command line
-
hf download hf://OpenSML/OpenSML-150M/evaluations/post_training_20260927.json
-
curl -L -o post_training_20260927.json https://huggingface.co/OpenSML/OpenSML-150M/resolve/main/evaluations/post_training_20260927.json
6.48 kB
| { | |
| "date": "2026-09-27", | |
| "training": false, | |
| "models": { | |
| "balanced-skills-1920": { | |
| "bundle": "sft_skill_balance_v1/step_0001920_628a0a65099b", | |
| "sha256": "4cac7f733e6ef63230e26e2aa66418bf0d9661ef55f46fa1e01f2ac602b17407", | |
| "results": { | |
| "multiple-choice": { | |
| "completed": 15428, | |
| "expected": 15428, | |
| "status": "complete", | |
| "tasks": { | |
| "arc_challenge": { | |
| "acc": 0.25341296928327645, | |
| "acc_norm": 0.3003412969283277, | |
| "completed": 1172, | |
| "expected": 1172 | |
| }, | |
| "arc_easy": { | |
| "acc": 0.5904882154882155, | |
| "acc_norm": 0.5707070707070707, | |
| "completed": 2376, | |
| "expected": 2376 | |
| }, | |
| "hellaswag": { | |
| "acc": 0.30999800836486757, | |
| "acc_norm": 0.34534953196574386, | |
| "completed": 10042, | |
| "expected": 10042 | |
| }, | |
| "piqa": { | |
| "acc": 0.6474428726877041, | |
| "acc_norm": 0.6430903155603918, | |
| "completed": 1838, | |
| "expected": 1838 | |
| } | |
| }, | |
| "training": false | |
| }, | |
| "ifeval": { | |
| "completed": 541, | |
| "expected": 541, | |
| "loose": { | |
| "instruction_accuracy": 0.2637889688249401, | |
| "instruction_correct": 220, | |
| "instruction_total": 834, | |
| "prompt_accuracy": 0.14417744916820702, | |
| "prompt_correct": 78, | |
| "prompt_total": 541 | |
| }, | |
| "status": "complete", | |
| "stop_counts": { | |
| "context_limit": 0, | |
| "eos": 539, | |
| "length": 2 | |
| }, | |
| "strict": { | |
| "instruction_accuracy": 0.2637889688249401, | |
| "instruction_correct": 220, | |
| "instruction_total": 834, | |
| "prompt_accuracy": 0.14417744916820702, | |
| "prompt_correct": 78, | |
| "prompt_total": 541 | |
| }, | |
| "training": false | |
| } | |
| } | |
| }, | |
| "constraint-completion-2048": { | |
| "bundle": "sft_constraint_completion_1920_v1/step_0002048_1267d9f4cf7a", | |
| "sha256": "e4e8a0de28e2141f6a580b49cc950aa64492c9f94d0d3aa98e6a26a2f439764b", | |
| "results": { | |
| "multiple-choice": { | |
| "completed": 15428, | |
| "expected": 15428, | |
| "status": "complete", | |
| "tasks": { | |
| "arc_challenge": { | |
| "acc": 0.25426621160409557, | |
| "acc_norm": 0.31399317406143346, | |
| "completed": 1172, | |
| "expected": 1172 | |
| }, | |
| "arc_easy": { | |
| "acc": 0.5871212121212122, | |
| "acc_norm": 0.5677609427609428, | |
| "completed": 2376, | |
| "expected": 2376 | |
| }, | |
| "hellaswag": { | |
| "acc": 0.31029675363473413, | |
| "acc_norm": 0.3472415853415654, | |
| "completed": 10042, | |
| "expected": 10042 | |
| }, | |
| "piqa": { | |
| "acc": 0.64689880304679, | |
| "acc_norm": 0.6381936887921654, | |
| "completed": 1838, | |
| "expected": 1838 | |
| } | |
| }, | |
| "training": false | |
| }, | |
| "ifeval": { | |
| "completed": 541, | |
| "expected": 541, | |
| "loose": { | |
| "instruction_accuracy": 0.29136690647482016, | |
| "instruction_correct": 243, | |
| "instruction_total": 834, | |
| "prompt_accuracy": 0.16820702402957485, | |
| "prompt_correct": 91, | |
| "prompt_total": 541 | |
| }, | |
| "status": "complete", | |
| "stop_counts": { | |
| "context_limit": 0, | |
| "eos": 532, | |
| "length": 9 | |
| }, | |
| "strict": { | |
| "instruction_accuracy": 0.28776978417266186, | |
| "instruction_correct": 240, | |
| "instruction_total": 834, | |
| "prompt_accuracy": 0.16635859519408502, | |
| "prompt_correct": 90, | |
| "prompt_total": 541 | |
| }, | |
| "training": false | |
| } | |
| } | |
| } | |
| }, | |
| "protocol": { | |
| "harness_reference_revision": "b954108c9baaaa934b4ad842033b31a97ee30816", | |
| "datasets": { | |
| "arc_easy": { | |
| "repo": "allenai/ai2_arc", | |
| "revision": "210d026faf9955653af8916fad021475a3f00453", | |
| "file": "ARC-Easy/test-00000-of-00001.parquet", | |
| "split": "test", | |
| "rows": 2376, | |
| "sha256": "4160597d618ae851c7eb04e281574f3f654776216ac6b6641588d64527b47177" | |
| }, | |
| "arc_challenge": { | |
| "repo": "allenai/ai2_arc", | |
| "revision": "210d026faf9955653af8916fad021475a3f00453", | |
| "file": "ARC-Challenge/test-00000-of-00001.parquet", | |
| "split": "test", | |
| "rows": 1172, | |
| "sha256": "62f03257e737aed263f55c6abf87c7bb0028a44a6bdd2a26eb1279eb42c1d1e9" | |
| }, | |
| "piqa": { | |
| "repo": "baber/piqa", | |
| "revision": "142f6d7367fd9877f0fb3b5734ea6a545f54cdd1", | |
| "file": "piqa_validation.parquet", | |
| "split": "validation", | |
| "rows": 1838, | |
| "sha256": "2b17ee1215b1ceb2ddabf4c90a05c5a8b833bef9d207960f57d111c43a2080f3" | |
| }, | |
| "hellaswag": { | |
| "repo": "Rowan/hellaswag", | |
| "revision": "218ec52e09a7e7462a5400043bb9a69a41d06b76", | |
| "file": "data/validation-00000-of-00001.parquet", | |
| "split": "validation", | |
| "rows": 10042, | |
| "sha256": "899813071e1e95efafec90f856e1987d2150fa4d020fc005df6962c259f660cd" | |
| }, | |
| "ifeval": { | |
| "repo": "google/IFEval", | |
| "revision": "966cd89545d6b6acfd7638bc708b98261ca58e84", | |
| "file": "ifeval_input_data.jsonl", | |
| "split": "train", | |
| "rows": 541, | |
| "sha256": "6a85310ca8ce15eff755aa08a3a4ff931c7e273e7515ebb3c492ea85fd8288f2" | |
| } | |
| }, | |
| "multiple_choice_normalization": "candidate log-likelihood divided by candidate character length", | |
| "few_shot_examples": 0, | |
| "ifeval": { | |
| "prompt_template": "User: {prompt}\nAssistant:", | |
| "max_context": 2048, | |
| "max_new_tokens": 1280, | |
| "greedy": true, | |
| "repetition_penalty": 1 | |
| } | |
| }, | |
| "limitations": [ | |
| "Custom MLX evaluator, not an official leaderboard submission.", | |
| "Public suites repeatedly inspected during development, not an untouched final test.", | |
| "No full contamination audit or statistical-significance claim.", | |
| "Instruction compliance does not establish correct content or conversational quality." | |
| ] | |
| } | |