Hugging Face's logo

Nairod785
/
Confucius4-R2T2-Q4_K_M-GGUF

Automatic Speech Recognition
GGUF
transcribe.cpp
asr
streaming
real-time
low-latency
quantized
q4_k_m
qwen3-asr
confucius4
r2t2
audio.cpp
Model card Files
xet
 Community
1
Nairod785 commited on
Commit
89e23a8
·
verified ·
1 Parent(s): 008a6fe

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +12 -8
README.md CHANGED
@@ -86,22 +86,26 @@ This release introduces an **asymmetric per-tensor quantization recipe** that br
86
 
87
  ---
88
 
89
- ## Comprehensive Quantization Report
90
 
91
- For the full ablation study, complete empirical test tables across 20+ quantization arms, failure mode analyses, speed benchmarks, and conversion reproduction instructions:
92
 
93
- 👉 **Read the full report in [QUANTIZATION.md](QUANTIZATION.md)**
 
 
94
 
95
  ---
96
 
97
  ## Files
98
 
99
- | File | Quantization | Size | Description |
100
  |---|---|---|---|
101
- | `r2t2-q4_k_m.gguf` | Q4_K_M | 1.187 GB | High-performance floor quantization; passes multilingual validation |
102
- | `QUANTIZATION.md` | — | — | Detailed technical report, ablation study, and benchmarks |
103
- | `NOTICE` | — | — | Attribution and derivative work disclaimers |
104
- | `LICENSE` / `LICENSE_zh` | — | — | NetEase Youdao Model Use License Agreement |
 
 
105
 
106
  ---
107
 
 
86
 
87
  ---
88
 
89
+ ## Technical Documentation & Reports
90
 
91
+ Detailed technical documentation and empirical logs accompanying this model:
92
 
93
+ - 📄 **[QUANTIZATION.md](QUANTIZATION.md)** — Empirical quantization report, 20+ arm evaluation table, failure mode analysis (cliff vs drift), speed benchmarks, and reproduction recipe.
94
+ - 📄 **[PORTING.md](PORTING.md)** — Architecture analysis, porting specifications, loader adaptation, streaming state machine, and chunk-size contract.
95
+ - 📄 **[QUANTIZATION_ARMS.md](QUANTIZATION_ARMS.md)** — The general per-tensor quantization arms methodology, `--keep-type` semantics, census asserting, and lessons learned.
96
 
97
  ---
98
 
99
  ## Files
100
 
101
+ | File | Type | Size | Description |
102
  |---|---|---|---|
103
+ | `r2t2-q4_k_m.gguf` | Model | 1.187 GB | High-performance floor quantization; passes multilingual validation |
104
+ | `QUANTIZATION.md` | Doc | — | Detailed technical report, ablation study, and benchmarks |
105
+ | `PORTING.md` | Doc | — | Porting and parity technical specification |
106
+ | `QUANTIZATION_ARMS.md` | Doc | — | General per-tensor quantization arms methodology |
107
+ | `NOTICE` | Legal | — | Attribution and derivative work disclaimers |
108
+ | `LICENSE` / `LICENSE_zh` | Legal | — | NetEase Youdao Model Use License Agreement |
109
 
110
  ---
111