Lemma
Collection
A lemma is "something assumed" — an intermediate theorem on the path to a larger proof, or a heading that signals the subject of what follows. • 16 items • Updated
Lemmy — Gemma 4 26B A4B MoE — MLX bf16 (full precision)
The Mixture-of-Experts member of the Lemma model family by Lethean. An EUPL-1.2 fork of Gemma 4 26B A4B with the Lethean Ethical Kernel (LEK) merged into the weights.
This repo hosts the MLX bf16 (full precision) build for native Apple Silicon inference via mlx-lm and mlx-vlm. For the GGUF playground (Ollama, llama.cpp) see lthn/lemmy. For the unmodified Google base see LetheanNetwork/lemmy.
Family
| Repo | Format | Bits |
|---|---|---|
lthn/lemmy |
GGUF multi-quant | Q4_K_M → BF16 |
lthn/lemmy-mlx |
MLX | 4-bit |
lthn/lemmy-mlx-8bit |
MLX | 8-bit |
lthn/lemmy-mlx-bf16 |
MLX | bf16 |
License
EUPL-1.2. See Gemma Terms of Use for upstream base model terms.
- Downloads last month
- 445
Model size
25B params
Tensor type
BF16
·
Hardware compatibility
Log In to add your hardware
Quantized