GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
textilelabs
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
textilelabs
Model Tree
Input Modalities
Output Modalities
Supported Functionality
| Loom Atom | keyword baseline | |
|---|---|---|
| held-out human utterances (1,860) | 96.0% | 64.9% |
| SNIPS — never seen in training (700) | 96.6% | 59.0% |
The second row is the one that matters. SNIPS played no part in training and the score does not drop — so this is not memorised phrasings. The keyword baseline is a hand-written list of ~75 tool-ish words scored on the identical splits.
Both test sets are balanced, so chance is 50%.
| parameters | ratio | |
|---|---|---|
| Loom Weave 2 | 59,650,000 | 2,664× |
| Loom Spark 2 | 19,867,008 | 887× |
| Loom Router 1 | 1,435,040 | 64× |
| Loom Atom | 22,392 | 1× |
57 KB. Small enough to embed as a byte array in a header file, and it runs in well under a millisecond on a CPU.
The cheapest useful decision in an agent stack: should this request touch a tool at all?
Put it in front of everything. If it says <none>, you have saved a retrieval call, a
router call, and possibly a large-model call — for the cost of a 57 KB matrix multiply. If
it says <tool>, hand off to something that decides which tool
(Loom Router 1 does that in one token
across 17 routes).
It is not a chat model, a router, or a classifier of intent. It answers one binary question and nothing else.
A full ladder was trained, four minutes per rung, identical data:
| params | dim | layers | held-out | SNIPS |
|---|---|---|---|---|
| 86,640 | 48 | 3 | 93.0% | 95.9% |
| 26,976 | 32 | 2 | 95.6% | 93.6% |
| 22,392 | 24 | 2 | 96.0% |
Depth matters more than width. Narrowing from 24 to 16 dimensions cost about 3 points. Dropping from two layers to one cost ten. One attention layer can notice keywords; two can combine a keyword with its context. The floor is a layer count, not a parameter count.
Every rung beats the keyword baseline — even 4,812 parameters, by 17 points.
One honest note: the 86,640-parameter model scores lowest on held-out data because every rung got the same four minutes, and it completed 3,006 optimiser steps against 24d2L's 10,809. It is under-trained, not worse. Do not read this table as "smaller is better".
Questions about the user personally — "what is my sister's name", "what did I have for
breakfast" — are the hard case. They need no tool (no tool can answer them), but they look
like lookups. Atom gets some right and some wrong; treat <tool> on a first-person
question as unreliable.
bash
ollama run hf.co/textilelabs/Loom-Atom "whats the weather in leeds"# <tool>
The template and params files in this repo are read automatically. params pins
temperature: 0 and num_predict: 1 — one token, deterministic.
python
import torchfrom transformers import AutoTokenizer, AutoModelForCausalLMtok = AutoTokenizer.from_pretrained("textilelabs/Loom-Atom")model = AutoModelForCausalLM.from_pretrained("textilelabs/Loom-Atom").eval()pair = torch.tensor([tok.convert_tokens_to_ids("<tool>"),tok.convert_tokens_to_ids("<none>")])def needs_tool(message: str) -> bool:p = f"<user>\n{message.strip()}\n<|eot|>\n<loom>\n"ids = tok(p, return_tensors="pt", add_special_tokens=False).input_idswith torch.no_grad():logits = model(input_ids=ids).logits[0, -1]# decide only between the two legal answersreturn bool(logits[pair].argmax() == 0)needs_tool("whats the weather in leeds") # Trueneeds_tool("i had a rough day") # False
Prompt format is exact: <user>\n{message}\n<|eot|>\n<loom>\n.
markdown
config.json / model.safetensors the model — 67 KBtokenizer.json / tokenizer_config.json custom BPE tokenizer, 512 tokensloom-atom-f16.gguf 57 KB, for Ollama / llama.cpptemplate / params read automatically by `ollama run hf.co/...`Modelfile for building locally
Real human utterances from two openly licensed corpora, relabelled to a single bit:
clinc/oos-eval (CC BY 3.0)15,502 utterances, balanced 50/50 by downsampling the majority class. A small procedurally generated slice written by Textile Labs covers "no tool needed" cases that public assistant corpora do not contain — chit-chat, ambiguity, and questions only the user can answer.
Both licences require attribution; this section satisfies that and must be kept with any redistribution.
Model: MIT. Training data retains its original licences and attribution as above.
| 9,392 | 16 | 2 | 93.4% | 94.1% |
| 6,800 | 16 | 1 | 83.0% | 84.4% |
| 4,812 | 12 | 1 | 82.2% | 81.9% |