GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
m87techlabs
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
m87techlabs
Model Tree
Input Modalities
Output Modalities
Supported Functionality
Generation starts from the prompt and stops at <eos> (token ID 1). An example prompt, exactly
as the evaluation used it:
text
% a reel in D major, 2/2 timeServe it and ask over HTTP. A copy downloaded from the Hugging Face Hub goes in
$SLM_HOME/models/abc-folk/3/ first; that is where slm serve looks for abc-folk:3.
bash
uv run slm serve --model abc-folk:3curl -s localhost:8000/generate -H 'content-type: application/json' \ -d '{"prompt": "% a reel in D major, 2/2 time\n", "seed": 0}'Or load it with Hugging Face transformers; the directory is a standard LlamaForCausalLM
checkpoint:
python
from transformers import AutoModelForCausalLM, AutoTokenizerpath = "$SLM_HOME/models/abc-folk/3" # or its Hugging Face Hub repo ID, if it was published theretok = AutoTokenizer.from_pretrained(path)model = AutoModelForCausalLM.from_pretrained(path)| architecture | LlamaForCausalLM (RMSNorm, RoPE, SwiGLU, tied embeddings) |
| parameters | 864,256 (853,120 non-embedding) |
| vocabulary | 87 tokens |
| context | 512 tokens |
| weights | model.safetensors, fp32, 3.5 MB |
| run | run-9b890db6dced (abc-baseline-sft), checkpoint best at step 1,000 |
| tokens seen (this stage) | 16,384,000 |
| validation loss | 1.1092 |
| GPU-hours (this stage) | 0.006 |
| tokenizer | tk-1030590a89ca |
| code | slmkit c34aed6 |
Full lineage back to the raw data: uv run slm lineage abc-folk:3.
From eval report ev-0ddc32e2a9: 3 sampling seeds × 200 samples, prompted with the project's sft prompts. Mean ± standard deviation across seeds.
| metric | score |
|---|---|
| ended | 0.982 ± 0.006 |
| length | 259.345 ± 4.707 |
| plays | 0.730 ± 0.048 |
| bar_accuracy | 0.851 ± 0.023 |
| ends_on_tonic | 0.340 ± 0.026 |
| novelty | 0.998 ± 0.001 |
Metric definitions: the project's README and docs/concepts/evaluation.md.
Checked before it was published, over 94 tokens: slmkit reloads the files with max
|Δlogit| = 0.0e+00 (identical). transformers 5.17.0 loads them with max |Δlogit| = 0.0e+00 and identical token IDs.
Trained on 4,248 tunes in ABC notation from three public-domain tune books: Ryan's Mammoth
Collection (1883), O'Neill's Music of Ireland (1903) and Aird's Selection of Airs (1778–1803), as
transcribed by volunteers and shipped in the music21
package (BSD-3-Clause).
One open question, stated plainly. The printed books are public domain, but the transcription files carry no licence of their own. Faithful transcriptions of public-domain scores are generally not treated as new creative work, but that is an assumption, not a grant. If a transcriber or rights-holder objects, the collection is removed, the model retrained, and this version withdrawn. The Session's data was deliberately not used: its licence forbids training language models.
The weights are released under Apache-2.0, the same licence as slmkit.
More detail: projects/abc_music/README.md in the slmkit repository.
<unk>, which the model learned nothing about. slm serve reports how many prompt tokens that
affected.uv run slm serve --model abc-folk:3curl -s localhost:8000/generate -H 'content-type: application/json' \ -d '{"prompt": "% a reel in D major, 2/2 time\n", "seed": 0}'from transformers import AutoModelForCausalLM, AutoTokenizerpath = "$SLM_HOME/models/abc-folk/3" # or its Hugging Face Hub repo ID, if it was published theretok = AutoTokenizer.from_pretrained(path)model = AutoModelForCausalLM.from_pretrained(path)