ornith-ai/Ornith-1.5-9B model card
repo
Fields
id
sources/hf-ornith-1-5-9b-cardPublisher
Ornith AI
Kind
repo
Published
2026-08-18
Accessed
2026-10-18
notes
Fetched 2026-10-18. Weights repo created 2026-08-18 (HF API createdAt 2026-08-18T06:20:12Z). Card describes a dense ~9B reasoning model, 262,144 native context, MIT, about 19 GB in BF16, served on one 80 GB GPU in the card's own recipe. The card's own benchmark table is a vendor claim, averaged over five runs: SWE-bench Verified 70.6, GPQA Diamond 86.4, Terminal-Bench 2.1 (Terminus-2) 46.2. The HF eval-results widget repeats SWE-bench Verified 70.6 and GPQA Diamond 86.4. HF safetensors metadata reports 9,653,104,368 BF16 parameters.
Notes
Fetched 2026-10-18. Weights repo created 2026-08-18 (HF API createdAt 2026-08-18T06:20:12Z). Card describes a dense ~9B reasoning model, 262,144 native context, MIT, about 19 GB in BF16, served on one 80 GB GPU in the card's own recipe. The card's own benchmark table is a vendor claim, averaged over five runs: SWE-bench Verified 70.6, GPQA Diamond 86.4, Terminal-Bench 2.1 (Terminus-2) 46.2. The HF eval-results widget repeats SWE-bench Verified 70.6 and GPQA Diamond 86.4. HF safetensors metadata reports 9,653,104,368 BF16 parameters.
Cross-references
This record references no other records.