Luigi Claude Sonnet 4.5 commited on
Commit
bc08390
Β·
1 Parent(s): 1f857e7

Reorder models by total parameter count (smallest to largest)

Browse files

Correct order:
100M β†’ 270M β†’ 300M β†’ 350M β†’ 350M β†’ 500M β†’ 500M β†’ 600M β†’ 1.5B β†’ 1.7B β†’ 2.6B β†’ 4B β†’ 7B

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Files changed (1) hide show
  1. app.py +26 -26
app.py CHANGED
@@ -30,19 +30,6 @@ MAX_USABLE_CTX = 32768
30
 
31
  # Available models registry - ordered by parameter count (smallest to largest)
32
  AVAILABLE_MODELS = {
33
- "granite4_tiny_q3": {
34
- "name": "Granite 4.0 Tiny 7B (128K Context)",
35
- "repo_id": "unsloth/granite-4.0-h-tiny-GGUF",
36
- "filename": "*Q3_K_M.gguf",
37
- "max_context": 131072,
38
- "supports_toggle": False,
39
- "inference_settings": {
40
- "temperature": 0.0,
41
- "top_p": 1.0,
42
- "top_k": 0,
43
- "repeat_penalty": 1.1,
44
- },
45
- },
46
  "falcon_h1_100m": {
47
  "name": "Falcon-H1 100M",
48
  "repo_id": "mradermacher/Falcon-H1-Tiny-Multilingual-100M-Instruct-GGUF",
@@ -152,19 +139,6 @@ AVAILABLE_MODELS = {
152
  "repeat_penalty": 1.05,
153
  },
154
  },
155
- "lfm2_2_6b_transcript": {
156
- "name": "LFM2 2.6B Transcript",
157
- "repo_id": "mradermacher/LFM2-2.6B-Transcript-GGUF",
158
- "filename": "*Q4_K_M.gguf",
159
- "max_context": 32768,
160
- "supports_toggle": False,
161
- "inference_settings": {
162
- "temperature": 0.3,
163
- "top_p": 0.9,
164
- "top_k": 40,
165
- "repeat_penalty": 1.1,
166
- },
167
- },
168
  "qwen3_1.7b_q4": {
169
  "name": "Qwen3 1.7B Q4",
170
  "repo_id": "unsloth/Qwen3-1.7B-GGUF",
@@ -178,6 +152,19 @@ AVAILABLE_MODELS = {
178
  "repeat_penalty": 1.05,
179
  },
180
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
181
  "qwen3_4b_thinking_q3": {
182
  "name": "Qwen3 4B Thinking (256K Context)",
183
  "repo_id": "unsloth/Qwen3-4B-Thinking-2507-GGUF",
@@ -191,6 +178,19 @@ AVAILABLE_MODELS = {
191
  "repeat_penalty": 1.0,
192
  },
193
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  }
195
 
196
  DEFAULT_MODEL_KEY = "qwen3_600m_q4"
 
30
 
31
  # Available models registry - ordered by parameter count (smallest to largest)
32
  AVAILABLE_MODELS = {
 
 
 
 
 
 
 
 
 
 
 
 
 
33
  "falcon_h1_100m": {
34
  "name": "Falcon-H1 100M",
35
  "repo_id": "mradermacher/Falcon-H1-Tiny-Multilingual-100M-Instruct-GGUF",
 
139
  "repeat_penalty": 1.05,
140
  },
141
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
142
  "qwen3_1.7b_q4": {
143
  "name": "Qwen3 1.7B Q4",
144
  "repo_id": "unsloth/Qwen3-1.7B-GGUF",
 
152
  "repeat_penalty": 1.05,
153
  },
154
  },
155
+ "lfm2_2_6b_transcript": {
156
+ "name": "LFM2 2.6B Transcript",
157
+ "repo_id": "mradermacher/LFM2-2.6B-Transcript-GGUF",
158
+ "filename": "*Q4_K_M.gguf",
159
+ "max_context": 32768,
160
+ "supports_toggle": False,
161
+ "inference_settings": {
162
+ "temperature": 0.3,
163
+ "top_p": 0.9,
164
+ "top_k": 40,
165
+ "repeat_penalty": 1.1,
166
+ },
167
+ },
168
  "qwen3_4b_thinking_q3": {
169
  "name": "Qwen3 4B Thinking (256K Context)",
170
  "repo_id": "unsloth/Qwen3-4B-Thinking-2507-GGUF",
 
178
  "repeat_penalty": 1.0,
179
  },
180
  },
181
+ "granite4_tiny_q3": {
182
+ "name": "Granite 4.0 Tiny 7B (128K Context)",
183
+ "repo_id": "unsloth/granite-4.0-h-tiny-GGUF",
184
+ "filename": "*Q3_K_M.gguf",
185
+ "max_context": 131072,
186
+ "supports_toggle": False,
187
+ "inference_settings": {
188
+ "temperature": 0.0,
189
+ "top_p": 1.0,
190
+ "top_k": 0,
191
+ "repeat_penalty": 1.1,
192
+ },
193
+ },
194
  }
195
 
196
  DEFAULT_MODEL_KEY = "qwen3_600m_q4"