Downloads · 30 days
6
7% of all-time downloads
Jnx03/kanitakorn-r1d-qwen14b-v11-dare-edu
kanitakorn-r1d-qwen14b-v11-dare-edu is a text generation model from Jnx03. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as mit.
Best-performing variant of the Kanitakorn Thai-Education LLM campaign (2026-05-23).
Downloads · 30 days
6
7% of all-time downloads
All-time downloads
85
Public
Parameters
14.8B
29.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors29.5 GB · 100%
From the Hugging Face model README
Best-performing variant of the Kanitakorn Thai-Education LLM campaign (2026-05-23).
A 14B-parameter Thai education AI built on deepseek-ai/DeepSeek-R1-Distill-Qwen-14B via mergekit DARE-linear with the SCB10X layer-asymmetric recipe (arXiv:2502.09056). Designed for Thai students from grade 1 (P1) through grade 12 (M6).
| Benchmark | Score | Method | Target |
|---|---|---|---|
| ThaiExam (cons@8) | 56.99% | 322/565 | 85 |
| xCOPA-TH (cons@8) | 79.00% | 79/100 | 80 |
| Belebele-TH (cons@4) | 84.33% ✅ | 759/900 | 80 |
| MATH500 (cons@4) | 64.80% ✅ | 324/500 | 60 |
| AIME24 (cons@8) | 13.33% | 4/30 | 25 (route to base R1) |
Two-model routing recommended: this model for Thai-language tasks (Belebele, ThaiExam, OpenThaiEval, IFEval-TH), base DeepSeek-R1-Distill-Qwen-14B for AIME24-style hard math (base scores 30% on AIME24 cons@8).
merge_method: dare_linear
dtype: bfloat16
tokenizer_source: base
base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
slices:
- sources:
- model: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
layer_range: [0, 32]
parameters: {weight: 0.75, density: 0.7}
- model: Jnx03/kanitakorn-r1d-qwen14b-v10-merged-fp16-edu
layer_range: [0, 32]
parameters: {weight: 0.25, density: 0.7}
- sources:
- model: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
layer_range: [32, 48]
parameters: {weight: 0.4, density: 0.7}
- model: Jnx03/kanitakorn-r1d-qwen14b-v10-merged-fp16-edu
layer_range: [32, 48]
parameters: {weight: 0.6, density: 0.7}
The v10-merged-fp16-edu source is a dense merge of R1-Distill-14B + v10 LoRA (a Thai-SFT adapter trained on 6209 records of P1-M6 curriculum + ThaiExam-style + OpenThaiEval).
Primary audience: Thai students (P1-M6, grades 1-12) and teachers.
Capabilities:
from vllm import LLM, SamplingParams
llm = LLM(
model="Jnx03/kanitakorn-r1d-qwen14b-v11-dare-edu",
dtype="bfloat16", max_model_len=4096, gpu_memory_utilization=0.90,
)
sampling = SamplingParams(n=8, temperature=0.6, top_p=0.95, max_tokens=2048)
prompt = "..." # Thai prompt + suffix asking for คำตอบคือ (X) format
outputs = llm.generate([prompt], sampling)
# Majority vote across n=8 samples for ThaiExam-style MCQ
If you use this model, please cite the SCB10X recipe:
Pipatanakul, K., et al. Adapting Language-Specific LLMs to a Reasoning Model in One Day via Model Merging. arXiv:2502.09056, 2025.
And the campaign log: https://github.com/JNX03/kanitakorn-research
License: MIT (inherited from DeepSeek-R1-Distill-Qwen-14B)