bionemo-kermt-finetune: worth installing? Reported results

Reported results for bionemo-kermt-finetune (NVIDIA/skills) from NVIDIA, one row per receipt, in plain words.

Skill skills/bionemo-kermt-finetune in NVIDIA/skills.

One row per receipt. No figure on this page combines two rows.

Reported by: Reported by NVIDIA in its benchmark card
Model: Claude Opus 4.8 (aws/anthropic/bedrock-claude-opus-4-8, as the source names it)
Harness: claude-code, version not stated
Date: 2026-09-15
Tasks and runs: 5 tasks, as printed
Result: NVIDIA reported a score of 84.9% overall and no uplift figure (3 attempts per task). Overall: 84.9% — baseline ran, but no comparable score was available; uplift unavailable Security: 88.5% → 100.0% (+11.5 points) Correctness: 23.1% → 96.0% (+72.9 points) Discoverability: 97.5% — baseline ran, but no comparable score was available; uplift unavailable Effectiveness: 18.9% → 51.0% (+32.1 points) Efficiency: 79.8% — baseline ran, but no comparable score was available; uplift unavailable
Tokens: 1,624,153 tokens with the skill and 5,610,370 without, over all cases, as printed; the card gives no cost
Freshness: Reported on Opus 4.8. Registered opus models released after this run, as of the build date: Opus 5.5.
Skill commit: 14a98ae278cec7e867695189d0f4003c08e4f765
Receipt: kermt-finetune-aws-anthropic-bedrock-claude-opus-4-8-imported-nvidia-cards-claude-code-2026-09-15-fc4b07b20f8b.json

Reported by: Reported by NVIDIA in its benchmark card
Model: GPT-5.5 (openai/openai/gpt-5.5, as the source names it)
Harness: codex, version not stated
Date: 2026-09-15
Tasks and runs: 5 tasks, as printed
Result: NVIDIA reported a score of 69.5% overall and no uplift figure (3 attempts per task). Overall: 69.5% — baseline ran, but no comparable score was available; uplift unavailable Security: 27.3% → 71.4% (+44.1 points) Correctness: 78.2% → 91.4% (+13.2 points) Discoverability: 79.2% — baseline ran, but no comparable score was available; uplift unavailable Effectiveness: 30.7% → 35.7% (+5.0 points) Efficiency: 69.7% — baseline ran, but no comparable score was available; uplift unavailable
Tokens: 1,907,233 tokens with the skill and 6,598,000 without, over all cases, as printed; the card gives no cost
Freshness: Reported on GPT-5.5. Freshness unknown: the registry has no release date for it.
Skill commit: 14a98ae278cec7e867695189d0f4003c08e4f765
Receipt: kermt-finetune-openai-openai-gpt-5-5-imported-nvidia-cards-codex-2026-09-15-fc4b07b20f8b.json

Built on 2026-10-15 from committed receipts.