Instructions to use Eeshan082004/quotedesk-student with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Eeshan082004/quotedesk-student with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-3-270m-it") model = PeftModel.from_pretrained(base_model, "Eeshan082004/quotedesk-student") - Notebooks
- Google Colab
- Kaggle
Download base_eval.json from Eeshan082004/quotedesk-student: direct link, hf CLI and curl.
- Browser
- Download file 2.97 kB
-
https://huggingface.co/Eeshan082004/quotedesk-student/resolve/main/base_eval.json
- Command line
-
hf download hf://Eeshan082004/quotedesk-student/base_eval.json
-
curl -L -o base_eval.json https://huggingface.co/Eeshan082004/quotedesk-student/resolve/main/base_eval.json
2.97 kB
| { | |
| "system": "base", | |
| "model": "unsloth/gemma-3-270m-it (3-shot)", | |
| "created": "2026-10-01T03:18:27", | |
| "splits": { | |
| "exam": { | |
| "n": 150, | |
| "intent_accuracy": 0.5067, | |
| "json_validity_rate": 0.66, | |
| "item_precision": 0.1703, | |
| "item_recall": 0.1794, | |
| "item_f1": 0.1747, | |
| "quantity_accuracy": 0.3052, | |
| "part_number_accuracy": 0.2174, | |
| "compatible_with_accuracy": 0.0, | |
| "missing_info_precision": 0.0, | |
| "missing_info_recall": 0.0, | |
| "missing_info_f1": 0.0, | |
| "urgency_accuracy": 0.56, | |
| "exact_match_rate": 0.0067, | |
| "per_intent_accuracy": { | |
| "order_status": 0.0, | |
| "other": 0.0, | |
| "product_question": 0.5, | |
| "quote_request": 0.6875, | |
| "return_request": 0.23076923076923078 | |
| }, | |
| "per_intent_n": { | |
| "order_status": 16, | |
| "other": 11, | |
| "product_question": 14, | |
| "quote_request": 96, | |
| "return_request": 13 | |
| }, | |
| "latency_p50_s": 1.2609, | |
| "latency_p95_s": 1.8841, | |
| "escalation_rate": null, | |
| "cost_per_1k_usd": null, | |
| "wall_time_s": 0.0, | |
| "errors": 0 | |
| }, | |
| "exam_hard": { | |
| "n": 30, | |
| "intent_accuracy": 0.3667, | |
| "json_validity_rate": 0.4667, | |
| "item_precision": 0.2791, | |
| "item_recall": 0.2727, | |
| "item_f1": 0.2759, | |
| "quantity_accuracy": 0.6316, | |
| "part_number_accuracy": 0.0, | |
| "compatible_with_accuracy": 0.0, | |
| "missing_info_precision": 0.0, | |
| "missing_info_recall": 0.0, | |
| "missing_info_f1": 0.0, | |
| "urgency_accuracy": 0.6, | |
| "exact_match_rate": 0.0333, | |
| "per_intent_accuracy": { | |
| "order_status": 0.0, | |
| "product_question": 0.6666666666666666, | |
| "quote_request": 0.36 | |
| }, | |
| "per_intent_n": { | |
| "order_status": 2, | |
| "product_question": 3, | |
| "quote_request": 25 | |
| }, | |
| "latency_p50_s": 0.908, | |
| "latency_p95_s": 1.0846, | |
| "escalation_rate": null, | |
| "cost_per_1k_usd": null, | |
| "wall_time_s": 0.0, | |
| "errors": 0 | |
| } | |
| }, | |
| "pricing": { | |
| "teacher": { | |
| "model": "gemini-3.8-flash", | |
| "input_per_1m_usd": 0.75, | |
| "output_per_1m_usd": 3.75, | |
| "note": "Google's published paid-tier price (standard), thinking tokens billed as output; valid through 2026-12-31. Our calls go through the KIE.ai proxy; credits actually spent are logged separately.", | |
| "source": "https://ai.google.dev/gemini-api/docs/pricing", | |
| "verified": "2026-09-30", | |
| "label": "measured tokens x published paid-tier price" | |
| }, | |
| "student_cpu": { | |
| "basis": "Modal CPU container, 2 cores + 4 GiB: 2 x $0.0000131/core/s + 4 x $0.00000222/GiB/s", | |
| "usd_per_second": 3.51e-05, | |
| "source": "https://modal.com/pricing", | |
| "verified": "2026-09-30", | |
| "label": "estimate: inference seconds x container price (excludes idle time)" | |
| } | |
| }, | |
| "hardware": "Tesla T4", | |
| "fewshot_ids": [ | |
| "S00001", | |
| "S00003", | |
| "S00007" | |
| ] | |
| } |