Hugging Face's logo

lthn
/
lemma

Image-Text-to-Text
GGUF
gemma4
lemma
llama.cpp
ollama
multimodal
vision
audio
on-device
conversational
Eval Results
4-bit precision
Model card Files
xet
 Community
lemma

Commit History

tokenizer_config: fix response schema regex — tool_calls before content, add tool_response terminator (from google upstream 2026-04-10)
5bea3ed
verified

lthn commited on

docs: correct base_model lineage for HF model tree
95a8db5

Snider Virgil commited on

cleanup: drop mlx files, align with lemer gguf-playground pattern
733d141

Snider Virgil commited on

eval: migrate results/ → .eval_results/ canonical storage
5cc8b73

Snider Virgil commited on

eval: migrate to HF .eval_results/*.yaml + TIGER-Lab/MMLU-Pro
6825f72

Snider Virgil commited on

eval(8-PAC): first paired benchmark run (n=4, r=8, mmlu_pro)
6af5ec0

Snider Virgil commited on

docs: update README for LEK-merged Lemma family
6351dde

Snider Virgil commited on

feat: merge LEK into lemma weights
4f98107

Snider Virgil commited on

fix: pull Google's chat_template.jinja fix (E2B/E4B variant)
a7e86f2

Snider Cladius Maximus commited on

fix: replace LICENSE symlink with real EUPL-1.2 file
fb71500

Snider Cladius Maximus commited on

Upload config.json with huggingface_hub
1d0c931
verified

lthn commited on

Upload tokenizer_config.json with huggingface_hub
f380ba3
verified

lthn commited on

Upload config.json with huggingface_hub
c8950f2
verified

lthn commited on

Upload processor_config.json with huggingface_hub
9068a39
verified

lthn commited on

Upload config.json with huggingface_hub
d1fe7bd
verified

lthn commited on

Upload folder using huggingface_hub
46ef3f0
verified

lthn commited on

Replace bf16 with Q4 MLX — consumer-ready quantisation
585a8a4
verified

lthn commited on

Upload Lemma E4B bf16 (MLX format) — LEK-activated Gemma 4 E4B
e083172
verified

lthn commited on

initial commit
a628c01
verified

lthn commited on