gabrielgts

granite4-3b-ec-magento

README

License: apache-2.0

What it does

Trained on 56,253 instruction samples: 70% ECInstruct (generic e-commerce) + 30% synthetic Magento-schema data generated from the Magento Luma sample catalog (products fully disjoint between train and eval). Four task shapes:

Attribute extraction — product text (or a raw Magento custom_attributes payload) → JSON:

markdown
target attribute: size
product title: Puma Suede green sneakers size 43
→ [{"attribute": "size", "value": "43"}]

Absent attributes are reported as "None" rather than hallucinated.

Product QA — a question answered strictly from given product data. Relevance classification — query + product → graded relevance option (ESCI-style A–D). Relevance ranking — query + lettered product list → ranked letters (B,A,C).

Usage — adapter (unsloth / peft)

python
from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    "gabrielgts/granite4-3b-ec-magento", max_seq_length=2048, load_in_4bit=True)
FastLanguageModel.for_inference(model)

# No system message — the model was trained without one, and Granite's template
# injects nothing when no tools/documents are passed. Greedy decoding recommended.
messages = [{"role": "user", "content":
    "Extract the value of the target attribute from the given product information "
    "and output it as JSON. If the attribute is not present, output None as the value.\n\n"
    "target attribute: size\nproduct title: Puma Suede green sneakers size 43"}]
text = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
inputs = tokenizer(text=text, return_tensors="pt").to("cuda")
out = model.generate(**inputs, max_new_tokens=64, do_sample=False)
print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
# [{"attribute": "size", "value": "43"}]

Plain peft also works: PeftModel.from_pretrained(base_model, "gabrielgts/granite4-3b-ec-magento").

Usage — GGUF / Ollama

bash
hf download gabrielgts/granite4-3b-ec-magento --include 'gguf/*' --local-dir .
cd gguf && ollama create granite4-3b-ec-magento -f Modelfile
ollama run granite4-3b-ec-magento "target attribute: color ..."

Quantization note: the shipped GGUF uses --token-embedding-type q8_0 --output-tensor-type q8_0. Granite ties its embeddings over a 100,352-token vocabulary, and tied large-vocab embeddings degrade at plain 4-bit — producing correctly-formatted but wrong answers. If you requantize, keep the embeddings at q8_0 and verify answer content, not just JSON validity.

Training recipe

Table with columns: Parameter, Value
Parameter	Value
Method	QLoRA (4-bit NF4 base, bf16 compute) via Unsloth
LoRA	r=8, alpha=16, targets q/k/v/o/gate/up/down_proj
Trainable params	15.6M of 3,403M (0.46%)
Batch	1 × grad_accum 16 (effective 16), max_seq_length 2048
Optimizer / LR	paged_adamw_8bit, 2e-4 cosine, 1 epoch, seed 42
Data	56,253 samples: 39,377 ECInstruct + 16,876 Magento-synthetic
System prompt	none at train and eval

Evaluation

Greedy decoding, identical prompts across all five models; base models evaluated zero-shot with the same harness. All five were trained on the same frozen dataset (sha256 afb7e664…), so the base model is the only variable.

Magento held-out set (2,969 samples, 475 products never seen in training). Bold = best of the five.

Table with columns: Task · metric, Base, this model, ministral3-3b, phi4-mini, qwen3.5-4b, gemma4-e2b
Task · metric	Base	this model	ministral3-3b	phi4-mini	qwen3.5-4b	gemma4-e2b
Attribute extraction · F1	0.106	0.953	0.945	0.946	0.938	0.933
Attribute extraction · parse failures	21.5%	0.1%	0.1%	0.3%

ECInstruct held-out set (2,000 samples):

Table with columns: Task · metric, Base, this model, ministral3-3b, phi4-mini, qwen3.5-4b, gemma4-e2b
Task · metric	Base	this model	ministral3-3b	phi4-mini	qwen3.5-4b	gemma4-e2b
Answerability · accuracy	0.448	0.838	0.735	0.733	0.780	0.728
Attribute extraction · F1	0.000	0.659	0.654	0.616	0.646

Long-list ranking (500 held-out ESCI queries, 5–8 candidates with graded hard negatives; chance top-1 = 0.167). This set is much harder than the ranking task above, which is 62% two-candidate:

Table with columns: top-1, exact-match
	top-1	exact-match
Base (zero-shot)	0.070	0.000
this model	0.516	0.008

Read that 0.516 against its own baseline, not against the 0.655 above: the easy set is mostly binary A/B choices, so a high score there flatters the model. Ranking is this model's relative weak spot.

Limitations — read before relying on the numbers

The Magento eval is synthetic-on-synthetic. Eval tasks were generated with the same templates as the training data (products fully disjoint). It validly measures schema adherence — JSON format, Magento attribute vocabularies, the None-when-absent rule — but overstates production quality on real catalogs and real user queries.
The recipe was transplanted, not tuned per model. r=8 / 2e-4 / 1 epoch was validated on a different base. These results mean "best under this fixed recipe", not an intrinsic ranking of the base models.
The base model's low structured scores are dominated by format non-adherence; they understate its underlying capability, though prose output is itself a blocker for programmatic use.
English only; fine-tuned on rigid structured data — expect degraded general chat and multilingual ability versus the base model (drop the adapter to recover it).
Answer-generation F1 is low by construction (free-form text scored against a single reference); use it comparatively, not absolutely.
Use greedy decoding (do_sample=False / temperature 0) — that is how it was evaluated.

Provenance

Table

Training run	`granite4-3b-r8-mix56k-e1`
Adapter sha256	`770f30f9c6abc2e06ce1f19f7ebac6fa5d9cb1e0160be35f76e11a3165f04d5b`
Train set sha256	`afb7e664cda4490597c1914db6aff94933991a56d2175435666f8f6b7a726532` (mixture_train.jsonl, 56,253 rows)
Eval set sha256	`8feafdb2…` (magento_eval.jsonl) · `2583a61e…` (ecinstruct_eval.jsonl)
GGUF	adapter merged into `ibm-granite/granite-4.1-3b` bf16, Q4_K_M + q8_0 embeddings, sha256

References

Granite 4.1: ibm-granite/granite-4.1-3b
ECInstruct / eCeLLM: arXiv:2402.08831
Amazon ESCI (Shopping Queries): arXiv:2206.06588
QLoRA: arXiv:2305.14314

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

gabrielgts

Model Tree

Base

ibm-granite/granite-4.1-3b

Adapter

this model

Input Modalities

Text

Output Modalities