Text Generation
PEFT
Safetensors
English
lora
qwen2.5
sakthai
house-of-sak
tool-calling
function-calling
agent
qlora
conversational
assistant
Eval Results
Eval Results (legacy)
Instructions to use Nanthasit/sakthai-context-7b-tools with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Nanthasit/sakthai-context-7b-tools with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-7B-Instruct") model = PeftModel.from_pretrained(base_model, "Nanthasit/sakthai-context-7b-tools") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Upload .eval_results/health-sakthai-context-7b-tools-2026-08-01.yaml with huggingface_hub
Verified signaturefa76b7a -
Add eval result: metadata cron snapshot
Verified signatureb564410 -
health check 2026-08-01: issues_found
Verified signature7df9d2a -
chore: metadata cron eval for sakthai-context-7b-tools 20260801T042432Z
SakThai Cronc3edf3a -
Upload README.md with huggingface_hub
Verified signature2a67b91 -
fix: close malformed BibTeX howpublished/note block in README
Verified signature7f94d3c -
Upload README.md with huggingface_hub
Verified signature1df5e52
-
fix: remove trailing } from self-link in README
Verified signaturecfae210 -
chore(eval): add cron metadata eval for sakthai-context-7b-tools
SakThai7fc74da -
ROUND-6 REFRESH: family table to 2026-07-31 live counts (26 repos)
Verified signature170805a -
Upload README.md with huggingface_hub
Verified signature9e35f5a -
fix(maintain): clean malformed cross-links
Verified signatured4378bf -
cron: add eval result #1 for sakthai-context-7b-tools (metadata_cron)
Verified signature552dddd -
Update family count: 20 public models -> 22 public models
Verified signatured0da5a9 -
benchmark roundup 20260731_114345
Verified signature65460b4 -
Fix stale/incorrect eval results with real measured numbers
Verified signature39ce48c -
eval results on bench-v2 (scorer multiset-selection-v2)
Verified signatured673738 -
Audit fix: family collection count 20 public models / 15 datasets / 4 Spaces (39 items, bench-v3 joined)
Verified signature620720d -
Add health check report 2026-07-31
Verified signature5d17d06 -
Health check 2026-07-31: correct training-table numbers from training_metrics.json (0.798->0.513; eval relabeled as run-log-reported), fix 1.5b-tools size row 11->8.7 MB
Verified signature7e59847 -
Fix stale training-data row count: combined-v6 (1,378)->(2,003) [verified live via datasets-server]
Verified signaturee006991 -
Card: add Arguments Accuracy to model-index, dedicated Benchmarks table with verification status, and Citation (Qwen2.5 + fine-tune)
Verified signature4446eef -
audit: TTS row size 135→141 MB (matches actual 141,322,336 B and sibling cards)
Verified signature2b8053f -
fix(crosslinks): correct stale Coder Browser GGUF size 6.6->7.1 GB (actual 7,111,560,032 B); remove phantom private/internal dataset claim (combined-v8 & cycle-bench 404)
Verified signature535c221 -
cron: add eval result #1 for sakthai-context-7b-tools (metadata_cron)
Verified signaturee98f4ee -
Upload README.md with huggingface_hub
Verified signature9feda9b -
Upload README.md with huggingface_hub
Verified signature4503a26
-
Upload .eval_results/sakthai-bench-v2.yaml with huggingface_hub
Verified signaturefbee654 -
Update family table: add 2 missing models (1.5b-merged-v2, embedding), bump count 12→14
Verified signature3071f5b -
Cross-link: add 1.5b-tools-v2 to family table + update model count 11→12
Verified signaturee02265b -
fix: model count 12→11 (actual from HF API)
Verified signaturefca007a -
add: The Story Behind It section with origin narrative
SakThai Agent3cfdba9 -
fix: update model family count from 13 to 12
Verified signaturefcbc402 -
fix: correct model count 14→13, remove deprecated English embedding reference
Verified signaturedf60509 -
fix: align family count to 14 + add signature tagline
Verified signature1c16132 -
docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts
Verified signaturee6b50f5 -
docs: align card to lean standard — live downloads badge, single canonical family table, drop funnel sections and stale counts
Verified signature1ce9ae0
-
Fix dead v2 refs, update ecosystem counts to 12/8/3, add bench-v1/v2 datasets
Verified signatureb06cbf5 -
fix: replace broken shields.io/endpoint badge with working static badge (219 dl)
Verified signature9add6ce -
Remove eval_results derived from small internal spot checks, not benchmarks
Verified signature5b7fb89 -
Remove unearned model-index claims; state measured results or mark as unbenchmarked
Verified signature9d1cfd2 -
Overhaul: expanded ecosystem cross-promotion, fix counts, add Growing the ecosystem section
Verified signature8e8449f -
Upload .eval_results/bfcl.yaml with huggingface_hub
Verified signatureac0867a -
Delete .eval_results/bfcl-irrelevance.yaml with huggingface_hub
Verified signature04f06ff -
Delete .eval_results/bfcl-parallel.yaml with huggingface_hub
Verified signaturec14894e -
Delete .eval_results/bfcl-multiple.yaml with huggingface_hub
Verified signature82d0d56 -
Delete .eval_results/bfcl-simple.yaml with huggingface_hub
Verified signaturefff9880 -
Upload .eval_results/bfcl.yaml with huggingface_hub
Verified signature9a2dfd8 -
Upload .eval_results/bfcl-irrelevance.yaml with huggingface_hub
Verified signature917a4bc -
Upload .eval_results/bfcl-parallel.yaml with huggingface_hub
Verified signature66b8e73