CoolFace
Modelpublic

Jnx03/kanitakorn-v60-format-null-upper12-checkpoint-1

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face
0likes5downloads
Model Card

Jnx03/kanitakorn-v60-format-null-upper12-checkpoint-1

Kanitakorn 2026-06-13 campaign artifact.

  • —Base: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
  • —Version: v60-format-null-upper12-checkpoint-1
  • —Method: sft-lora-upper12
  • —Final-claim policy: one model, no BoN, no self-consistency, no ensemble, no model routing.
  • —Date: 2026-06-15

Scores

BenchmarkScoreTargetStatus
aime24_thpending>15.0pending
aime24pending>25.0pending
math500_thpending>56.0pending
math500pending>82.0pending
livecodebench_thpending>35.0pending
livecodebenchpending>60.0pending
openthaievalpending>80.0pending
hotpotqathenpending>46.0pending
instructionfollowingth_enpending>57.0pending
mtbenchth_enpending>85.0pending
thaiexampending>70.0pending
ifeval_thpending>82.0pending

Notes

Upper-layer-only LoRA from alpha0.90 on v58 format/null data. Single-model artifact; no BoN/self-consistency/routing.