Unverified50% confidenceBenchmarkExact time
On HumanEval, Keera's pretrained model achieved pass@1 of 3/164 (1.8%), and after fine-tuning 9/164 (5.5%).
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn the HumanEval pass@1 comparison, phi-1-small (350M parameters) reached 45%, while Keera fine-tuned (97.5M parameters) reached 5.5%.74% similarUnverifiedA 360M-parameter code model trained by the trifero.ai lab under Strife Technologies initially reported 26.8% pass@1 on HumanEval, and after removing instruction data overlapping with the benchmark, the true number was 22.6%.66% similarUnverifiedLoRA使可训练参数量通常仅为全参微调的0.1%-1%，可在单张RTX 3090/4090（显存24GB）上微调7B乃至13B级别的模型62% similarUnverifiedThe final parameter count of the model after training was 13,693,824.61% similarUnverifiedThe original Codex achieved a pass@1 score of 28.8% on the HumanEval benchmark, whereas using GPT-3 directly yielded only 0%.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1057719API
curl https://kongchang.com/api/v1/knowledge/claims/1057719MCP
get_claim(id=1057719)