Hugging Face's logo

majentik
/
Leanstral-RotorQuant-MLX-8bit

Text Generation
MLX
Safetensors
mistral3
rotorquant
kv-cache-quantization
8bit
weight-quantization
leanstral
lean4
formal-proofs
theorem-proving
quantized
apple-silicon
mistral
Mixture of Experts
8-bit precision
Model card Files
xet
 Community
majentik commited on
Commit
391ba0d
·
verified ·
1 Parent(s): d1dec42

Card accuracy pass 3: remove brand-comparison claim tables

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -90,7 +90,7 @@ upstream options described above (`-ctk/-ctv q8_0`, `OLLAMA_KV_CACHE_TYPE`).
90
  | Method | Prefill Speed | Decode Speed | Memory Savings | Reference |
91
  |---|---|---|---|---|
92
  | **TurboQuant** | Baseline | Baseline | High | [arXiv: 2504.19874](https://arxiv.org/abs/2504.19874) |
93
- | **RotorQuant** | different | 28% faster | High | [GitHub](https://github.com/scrya-com/rotorquant) |
94
 
95
  ## Memory Estimates
96
 
 
90
  | Method | Prefill Speed | Decode Speed | Memory Savings | Reference |
91
  |---|---|---|---|---|
92
  | **TurboQuant** | Baseline | Baseline | High | [arXiv: 2504.19874](https://arxiv.org/abs/2504.19874) |
93
+
94
 
95
  ## Memory Estimates
96