Naive-N0.5-Flash-JANGHT2.5
> ⚠️ Runtime: JANGHT needs a vMLX (Python) build that is not released yet. The bundle was built...
> ⚠️ Runtime: JANGHT needs a vMLX (Python) build that is not released yet. The bundle was built...
Built for vMLX — the MLX inference engine for Apple Silicon with mixed-precision JANG bundles, KV-cache quantization, and...
Spark-X2.5-4B at 3.41 GiB — the size-first bundle. Fastest of the pair, and the one to run if...
The standard profile. 8-bit attention/routers/shared experts, 4-bit routed experts. JANG mixed-precision affine (JANG4M, group size 64) conversion of...
JANGTQ (JANG TurboQuant) квантование zai-org/GLM-5.2 для MLX на кремнии Apple. TurboQuant применяет вращение Адамара со случайным знаком, норму...
JANGTQ (JANG TurboQuant) квантование zai-org/GLM-5.2 (744B-параметр glmmoedsa MoE) для MLX на кремнии Apple. TurboQuant применяет вращение Адамара со...
Tencent Hy3-preview — 102 GB on disk (down from ~557 GB BF16 source) — mixed-bit JANGTQK quantization on...
> Important: This model uses the JANGTQ (JANG TurboQuant) quantization format — an extreme-compression variant of JANG for...
INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED...
JANG-quantized Gemma-4 MoE for Apple Silicon. Created by Jinho Jang — eric@jangq.ai. Stock mlxlm picks up the multi-stop-token...