Hugging Face's logo

hitonet
/
hito-1.7b-GGUF

Text Generation
GGUF
English
qwen3
fine-tuned
hito
hitonet
reasoning
thinking
llama-cpp
ollama
conversational
Model card Files
xet
Community
hitonet commited on
Commit
9d93a28
·
verified ·
1 Parent(s): 9368769

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +38 -43
README.md CHANGED
@@ -17,21 +17,23 @@ pipeline_tag: text-generation
17
  base_model: hitonet/hito-1.7b
18
  ---
19
 
 
 
20
  # Hito 1.7B - GGUF
21
 
22
  ### Quantized versions for llama.cpp, Ollama, LM Studio, and more
23
 
24
- <p align="center">
25
- <a href="https://huggingface.co/hitonet/hito-1.7b">Original Model</a> •
26
- <a href="https://chat.hitonet.com">Web Chat</a> •
27
- <a href="https://platform.hitonet.com">API</a>
28
- </p>
29
 
30
  ---
31
 
32
- | Model License | Method License |
33
- |--------------|----------------|
34
- | Apache 2.0 ✅ | CC BY-NC-ND ⚠️ |
 
35
 
36
  ---
37
 
@@ -41,42 +43,44 @@ This repository contains **GGUF quantized versions** of [hitonet/hito-1.7b](http
41
 
42
  Hito is a 1.7B parameter model with **Nested Cognitive Reasoning** - structured, self-correcting thinking patterns that enable better accuracy and transparency.
43
 
 
 
44
  ---
45
 
46
- ## Quantization Quality Guide
47
 
48
- ### ⭐ Recommended (Best Quality)
49
 
50
  | File | Quant | Size | Quality | Notes |
51
  |------|-------|------|---------|-------|
52
- | **[hito-1.7b-Q4_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_K_M.gguf)** | Q4_K_M | 1.1 GB | ⭐ **Best** | Perfect balance - recommended for most users |
53
- | [hito-1.7b-Q5_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_K_M.gguf) | Q5_K_M | 1.2 GB | ⭐ Excellent | Slightly better quality |
54
  | [hito-1.7b-Q8_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q8_0.gguf) | Q8_0 | 1.8 GB | ⭐ Excellent | Highest quality quantization |
55
 
56
  ### ✅ Good Quality
57
 
58
- | File | Quant | Size | Quality |
59
- |------|-------|------|---------|
60
- | [hito-1.7b-Q4_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_0.gguf) | Q4_0 | 1.0 GB | ✅ Good |
61
- | [hito-1.7b-Q4_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_K_S.gguf) | Q4_K_S | 1.0 GB | ✅ Good |
62
- | [hito-1.7b-Q5_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_0.gguf) | Q5_0 | 1.2 GB | ✅ Good |
63
- | [hito-1.7b-Q5_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_K_S.gguf) | Q5_K_S | 1.2 GB | ✅ Good |
64
- | [hito-1.7b-Q6_K.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q6_K.gguf) | Q6_K | 1.4 GB | ✅ Excellent |
65
- | [hito-1.7b-F16.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-F16.gguf) | F16 | 3.3 GB | ✅ Reference |
66
 
67
- ### ⚠️ Lower Quality (Small Devices Only)
68
 
69
- | File | Quant | Size | Quality |
70
- |------|-------|------|---------|
71
- | [hito-1.7b-Q3_K_L.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_L.gguf) | Q3_K_L | 957 MB | ⚠️ Fair |
72
- | [hito-1.7b-Q3_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_M.gguf) | Q3_K_M | 896 MB | ⚠️ Fair |
73
- | [hito-1.7b-Q3_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_S.gguf) | Q3_K_S | 827 MB | ⚠️ Fair |
74
 
75
- ### ❌ Not Recommended
76
 
77
- | File | Quant | Size | Quality |
78
- |------|-------|------|---------|
79
- | [hito-1.7b-Q2_K.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q2_K.gguf) | Q2_K | 742 MB | ❌ Poor |
80
 
81
  ---
82
 
@@ -107,7 +111,7 @@ ollama run hito
107
  ./llama-cli -m hito-1.7b-Q4_K_M.gguf \
108
  -sys "You are Hito by Hitonet.com." \
109
  -p "What is your name?" \
110
- -n 256 --temp 0.7
111
  ```
112
 
113
  ### LM Studio
@@ -155,15 +159,6 @@ For full documentation, benchmarks, and training details, see the [main reposito
155
 
156
  ---
157
 
158
- ## Links
159
-
160
- - **Original Model:** [hitonet/hito-1.7b](https://huggingface.co/hitonet/hito-1.7b)
161
- - **Website:** [hitonet.com](https://hitonet.com)
162
- - **Chat:** [chat.hitonet.com](https://chat.hitonet.com)
163
- - **API:** [platform.hitonet.com](https://platform.hitonet.com)
164
-
165
- ---
166
-
167
- <p align="center">
168
- <b>Made with genuine curiosity by Hitonet</b>
169
- </p>
 
17
  base_model: hitonet/hito-1.7b
18
  ---
19
 
20
+ <div align="center">
21
+
22
  # Hito 1.7B - GGUF
23
 
24
  ### Quantized versions for llama.cpp, Ollama, LM Studio, and more
25
 
26
+ [![Original Model](https://img.shields.io/badge/Original_Model-ff6b35?style=for-the-badge)](https://huggingface.co/hitonet/hito-1.7b)
27
+ [![Website](https://img.shields.io/badge/hitonet.com-000000?style=for-the-badge&logo=globe&logoColor=white)](https://hitonet.com)
28
+ [![Chat](https://img.shields.io/badge/Try_Free_Chat-22c55e?style=for-the-badge&logo=chatbot&logoColor=white)](https://chat.hitonet.com)
29
+ [![API](https://img.shields.io/badge/API_Platform-3b82f6?style=for-the-badge&logo=swagger&logoColor=white)](https://platform.hitonet.com)
 
30
 
31
  ---
32
 
33
+ <img src="https://img.shields.io/badge/Model_Weights-Apache_2.0_(Open)-green?style=flat-square" alt="Model License"/>
34
+ <img src="https://img.shields.io/badge/NCR_Method-CC_BY--NC--ND_(Commercial_License_Required)-red?style=flat-square" alt="Method License"/>
35
+
36
+ </div>
37
 
38
  ---
39
 
 
43
 
44
  Hito is a 1.7B parameter model with **Nested Cognitive Reasoning** - structured, self-correcting thinking patterns that enable better accuracy and transparency.
45
 
46
+ For the original model (safetensors), training details, benchmarks, and full documentation, see the [main repository](https://huggingface.co/hitonet/hito-1.7b).
47
+
48
  ---
49
 
50
+ ## Available Quantizations
51
 
52
+ ### ⭐ Recommended
53
 
54
  | File | Quant | Size | Quality | Notes |
55
  |------|-------|------|---------|-------|
56
+ | **[hito-1.7b-Q4_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_K_M.gguf)** | Q4_K_M | 1.1 GB | ⭐ **BEST** | Perfect balance of size and quality |
57
+ | [hito-1.7b-Q5_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_K_M.gguf) | Q5_K_M | 1.2 GB | ⭐ Excellent | Slightly better than Q4_K_M |
58
  | [hito-1.7b-Q8_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q8_0.gguf) | Q8_0 | 1.8 GB | ⭐ Excellent | Highest quality quantization |
59
 
60
  ### ✅ Good Quality
61
 
62
+ | File | Quant | Size | Quality | Notes |
63
+ |------|-------|------|---------|-------|
64
+ | [hito-1.7b-Q4_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_0.gguf) | Q4_0 | 1.0 GB | ✅ Good | Legacy format, works well |
65
+ | [hito-1.7b-Q4_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q4_K_S.gguf) | Q4_K_S | 1.0 GB | ✅ Good | Smaller Q4 variant |
66
+ | [hito-1.7b-Q5_0.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_0.gguf) | Q5_0 | 1.2 GB | ✅ Good | Legacy 5-bit |
67
+ | [hito-1.7b-Q5_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q5_K_S.gguf) | Q5_K_S | 1.2 GB | ✅ Good | Smaller Q5 variant |
68
+ | [hito-1.7b-Q6_K.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q6_K.gguf) | Q6_K | 1.4 GB | ✅ Excellent | Near full quality |
69
+ | [hito-1.7b-F16.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-F16.gguf) | F16 | 3.3 GB | ✅ Reference | Full precision GGUF |
70
 
71
+ ### ⚠️ Low Quality (Not Recommended)
72
 
73
+ | File | Quant | Size | Quality | Notes |
74
+ |------|-------|------|---------|-------|
75
+ | [hito-1.7b-Q3_K_L.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_L.gguf) | Q3_K_L | 957 MB | ⚠️ Fair | May get stuck in thinking |
76
+ | [hito-1.7b-Q3_K_M.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_M.gguf) | Q3_K_M | 896 MB | ⚠️ Fair | Occasional issues |
77
+ | [hito-1.7b-Q3_K_S.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q3_K_S.gguf) | Q3_K_S | 827 MB | ⚠️ Fair | Noticeable quality loss |
78
 
79
+ ### ❌ Broken (Do Not Use)
80
 
81
+ | File | Quant | Size | Quality | Notes |
82
+ |------|-------|------|---------|-------|
83
+ | [hito-1.7b-Q2_K.gguf](https://huggingface.co/hitonet/hito-1.7b-GGUF/resolve/main/hito-1.7b-Q2_K.gguf) | Q2_K | 742 MB | ❌ Broken | Produces gibberish |
84
 
85
  ---
86
 
 
111
  ./llama-cli -m hito-1.7b-Q4_K_M.gguf \
112
  -sys "You are Hito by Hitonet.com." \
113
  -p "What is your name?" \
114
+ -n 256
115
  ```
116
 
117
  ### LM Studio
 
159
 
160
  ---
161
 
162
+ <div align="center">
163
+ <b>Made with genuine curiosity by Hitonet</b>
164
+ </div>