Stack-3.0-Omni-Nexus / benchmarks /hellaswag.json
Walid Sobhi
Add benchmark results: results_hellaswag_omni-nexus-alpha-q8_chat.json
7e3cf43 verified
Raw
History Blame Contribute Delete
191 Bytes
{
"benchmark": "hellaswag",
"model": "omni-nexus-alpha-q8",
"method": "chat-api (single generate, A/B/C/D pick)",
"accuracy": 0.5960963951404102,
"correct": 5986,
"total": 10042
}