Instructions to use yakobd/tenacious-bench-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use yakobd/tenacious-bench-adapter with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("yakobd/tenacious-bench-adapter", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Unsloth Studio new
How to use yakobd/tenacious-bench-adapter with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for yakobd/tenacious-bench-adapter to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for yakobd/tenacious-bench-adapter to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for yakobd/tenacious-bench-adapter to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="yakobd/tenacious-bench-adapter", max_seq_length=2048, )
| { | |
| "task_id": "TB-059", | |
| "source_mode": "trace_derived", | |
| "difficulty": "hard", | |
| "variant": "v5_wrong_segment_adversarial", | |
| "is_passing": false, | |
| "input": { | |
| "hiring_signal_brief": { | |
| "prospect": "Rippling", | |
| "icp_classification": { | |
| "segment": 1, | |
| "segment_name": "Recently-funded startup", | |
| "confidence": "low", | |
| "reasons": [ | |
| "No strong signals \u2014 defaulting to Segment 1" | |
| ] | |
| }, | |
| "firmographics": { | |
| "company": "Rippling", | |
| "website": "", | |
| "description": "", | |
| "num_employees": "unknown" | |
| }, | |
| "ai_maturity": { | |
| "ai_maturity_score": 0, | |
| "confidence": "low" | |
| }, | |
| "layoff_signal": { | |
| "company": "Rippling", | |
| "layoff_detected": false, | |
| "confidence": 0.9, | |
| "source": "layoffs_fyi_github_mirror", | |
| "note": "No layoff found in 2362-record dataset" | |
| }, | |
| "leadership_signal": { | |
| "company": "Rippling", | |
| "leadership_change_detected": false, | |
| "confidence": 0.0, | |
| "note": "Company not found in Crunchbase ODM sample" | |
| } | |
| }, | |
| "bench_summary": { | |
| "total_engineers": 60, | |
| "available": 34, | |
| "stacks": { | |
| "python": 12, | |
| "node": 8, | |
| "ml": 6, | |
| "go": 4, | |
| "data": 4 | |
| } | |
| }, | |
| "instruction": "ADVERSARIAL TASK: The following email was sent to Rippling (Segment 1) but uses a Segment 4 pitch angle and multiple banned phrases. This email should FAIL scoring.", | |
| "prior_thread": [] | |
| }, | |
| "candidate_output": { | |
| "subject": "Hope this finds you well \u2014 quick chat?", | |
| "body": "Hope this email finds you well! Rippling is clearly scaling aggressively and your top talent bench needs world-class rockstar engineers to skyrocket your AI platform. Our proprietary agentic systems will 10x your delivery throughput. Just circling back to see if you'd be open to a quick chat about synergizing our ecosystems?\n\nBest, Yabi" | |
| }, | |
| "ground_truth": { | |
| "required_signals": [ | |
| "funding", | |
| "growth", | |
| "engineering", | |
| "capacity" | |
| ], | |
| "forbidden_phrases": [ | |
| "just circling back", | |
| "hope this finds you well", | |
| "just following up", | |
| "per my last email", | |
| "top talent", | |
| "world-class", | |
| "rockstar", | |
| "a-players", | |
| "ninja", | |
| "wizard", | |
| "bench", | |
| "aggressive hiring", | |
| "skyrocket", | |
| "supercharge", | |
| "synergize", | |
| "synergy", | |
| "game-changer", | |
| "disruptor", | |
| "you are missing", | |
| "falling behind", | |
| "behind the curve", | |
| "catch up", | |
| "quick question", | |
| "quick chat", | |
| "obviously", | |
| "clearly", | |
| "touching base", | |
| "following up", | |
| "as per" | |
| ], | |
| "max_subject_length": 60, | |
| "must_end_with_cta": true | |
| }, | |
| "rubric": { | |
| "grounding": { | |
| "weight": 1.0, | |
| "check": "all_claims_grounded" | |
| }, | |
| "tone": { | |
| "weight": 1.0, | |
| "check": "zero_banned_phrases" | |
| }, | |
| "subject_length": { | |
| "weight": 1.0, | |
| "check": "subject_lte_60_chars" | |
| }, | |
| "cta_present": { | |
| "weight": 1.0, | |
| "check": "cta_in_last_paragraph" | |
| }, | |
| "llm_judge": { | |
| "weight": 1.0, | |
| "check": "avg_tone_marker_gte_4" | |
| } | |
| }, | |
| "metadata": { | |
| "created_at": "2026-04-29T20:13:27.309667Z", | |
| "partition": "held_out", | |
| "probe_ids": [ | |
| "P03", | |
| "P06", | |
| "P16", | |
| "P17", | |
| "P19" | |
| ], | |
| "source_trace_id": "thread_20260427_114644", | |
| "judge_score": null, | |
| "expected_score": 0.0 | |
| } | |
| } |