Downloads · 30 days
58
20% of all-time downloads
build-small-hackathon/mind-of-tashi-micro-sft-gguf
mind-of-tashi-micro-sft-gguf is a text generation model from build-small-hackathon. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
The fine-tuned GGUF the playable Hugging Face Space loads through llama.cpp — the Off the Grid + Llama Champion contract (no cloud API at runtime). Transformers source: build-small-hackathon/mind-of-tashi-micro-sft.
Downloads · 30 days
58
20% of all-time downloads
All-time downloads
286
Public
Repo size
1 GB
Likes
0
Public
Click a slice to open those files.
.gguf1 GB · 100%
From the Hugging Face model README
The fine-tuned GGUF the playable Hugging Face Space
loads through llama.cpp — the Off the Grid + Llama Champion contract
(no cloud API at runtime). Transformers source:
build-small-hackathon/mind-of-tashi-micro-sft.
David vs Goliath. Custom MoE loggenix-moe-0.4B-0.2A (~0.4B total /
~200M active per token), SFT'd to read an opponent and commit blind in an
English + Hindi/Sanskrit (IAST) code-switched register — 10–100× smaller
(active) than the API teachers it learned from.
| File | Size | Use |
|---|---|---|
mind-of-tashi-micro-sft-Q4_K_M.gguf | ~256 MB | deployed — small + fast |
mind-of-tashi-micro-sft-f16.gguf | ~786 MB | zero-loss reference |
chat_template.jinja (recent
checkpoints externalise the template; a missing one yields an empty-template
GGUF that loads but generates garbage).python convert_hf_to_gguf.py <ckpt> --outtype f16 → f16 GGUF.llama-cpp-python C binding (no cmake/MSVC):
import ctypes, llama_cpp
p = llama_cpp.llama_model_quantize_default_params()
p.ftype = 15 # LLAMA_FTYPE_MOSTLY_Q4_K_M
llama_cpp.llama_model_quantize(b"in-f16.gguf", b"out-Q4_K_M.gguf", ctypes.byref(p))
norm_topk_prob — required for llama.cppllama.cpp's qwen3moe graph hardcodes norm_w=true; the source checkpoint is
trained with norm_topk_prob=true to match. (See the SFT model card.)
<think> + {move,taunt}; ~19–20/20 bilingual.n_gpu_layers=99, n_ctx 4096):
cold load 3.2 s, TTFT ~97 ms (p50), decode ~450 t/s (p50), ~0.30 s/move,
peak VRAM < ~0.8 GB. CPU-only deploy is slower but fine for a 0.4B.from llama_cpp import Llama
llm = Llama.from_pretrained(
repo_id="build-small-hackathon/mind-of-tashi-micro-sft-gguf",
filename="mind-of-tashi-micro-sft-Q4_K_M.gguf",
n_ctx=4096, n_gpu_layers=0, # 0 = CPU (Space); 99 = full GPU offload
logits_all=True, # needed for the in-game Conviction Meter (token logprobs)
)
out = llm.create_chat_completion(messages=[
{"role": "system", "content": "<persona system prompt>"},
{"role": "user", "content": "<arena state + match history>"},
])
# -> "<think>...IAST-coded reasoning...</think>\n{\"move\": \"...\", \"taunt\": \"...\"}"
Game Space · self-play dataset · SFT model + GGUF (this) · OpenEnv gym ·
GRPO model + GGUF — all under build-small-hackathon/mind-of-tashi-*.