ai-expert armand0e commited on
Commit
6af5032
·
0 Parent(s):

Duplicate from TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill-GGUF

Browse files
.gitattributes ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ glm-4.7-flash-claude-4.5-opus.bf16.gguf filter=lfs diff=lfs merge=lfs -text
37
+ glm-4.7-flash-claude-4.5-opus.f16.gguf filter=lfs diff=lfs merge=lfs -text
38
+ glm-4.7-flash-claude-4.5-opus.q8_0.gguf filter=lfs diff=lfs merge=lfs -text
39
+ glm-4.7-flash-claude-4.5-opus.q6_k.gguf filter=lfs diff=lfs merge=lfs -text
40
+ glm-4.7-flash-claude-4.5-opus.q5_k_m.gguf filter=lfs diff=lfs merge=lfs -text
41
+ glm-4.7-flash-claude-4.5-opus.q4_k_m.gguf filter=lfs diff=lfs merge=lfs -text
42
+ glm-4.7-flash-claude-4.5-opus.q3_k_m.gguf filter=lfs diff=lfs merge=lfs -text
43
+ glm-4.7-flash-claude-4.5-opus.q3_k_s.gguf filter=lfs diff=lfs merge=lfs -text
44
+ glm-4.7-flash-claude-4.5-opus.iq4_nl.gguf filter=lfs diff=lfs merge=lfs -text
45
+ glm-4.7-flash-claude-4.5-opus.iq4_xs.gguf filter=lfs diff=lfs merge=lfs -text
46
+ glm-4.7-flash-claude-4.5-opus.iq3_xs.gguf filter=lfs diff=lfs merge=lfs -text
47
+ glm-4.7-flash-claude-4.5-opus.iq2_m.gguf filter=lfs diff=lfs merge=lfs -text
48
+ glm-4.7-flash-claude-4.5-opus.iq3_m.gguf filter=lfs diff=lfs merge=lfs -text
49
+ mmlu_subject_breakdown.png filter=lfs diff=lfs merge=lfs -text
50
+ results_bar_chart.png filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill
3
+ tags:
4
+ - text-generation-inference
5
+ - gguf
6
+ - llama.cpp
7
+ - unsloth
8
+ - glm4_moe_lite
9
+ license: apache-2.0
10
+ datasets:
11
+ - TeichAI/claude-4.5-opus-high-reasoning-250x
12
+ ---
13
+
14
+ # GLM 4.7 Flash x Claude 4.5 Opus (High Reasoning)
15
+
16
+ This model was trained on a small reasoning dataset of **Claude Opus 4.5**, with reasoning effort set to High.
17
+
18
+ - 🧬 Datasets:
19
+ - `TeichAI/claude-4.5-opus-high-reasoning-250x`
20
+
21
+ - 🏗 Base Model:
22
+ - `unsloth/GLM-4.7-Flash`
23
+
24
+ - ⚡ Use cases:
25
+ - Coding
26
+ - Science
27
+ - Deep Research
28
+
29
+ - ∑ Stats (Dataset)
30
+ - Costs: $ 52.30 (USD)
31
+ - Total tokens (input + output): 2.13 M
32
+
33
+ ---
34
+
35
+ ## How to run
36
+
37
+ For specific instructions/commands to serve this model locally using vLLM, SGLang, or transformers please see [the instructions from the original model's card](https://huggingface.co/unsloth/GLM-4.7-Flash#serve-glm-47-flash-locally)
38
+
39
+ For detailed instructions getting started with Llama.cpp please refer to [the unsloth guide](https://unsloth.ai/docs/models/glm-4.7-flash)
40
+
41
+ ### Sampling Parameters
42
+
43
+ z-ai recommends the following sampling parameters for this model:
44
+
45
+ | Default Settings (Most Tasks) | Terminal Bench, SWE Bench Verified |
46
+ | ------------------------------------------------------------------ | ------------------------------------------------------------------ |
47
+ | **temperature = 1.0** | **temperature = 0.7** |
48
+ | **top_p = 0.95** | **top_p = 1.0** |
49
+ | repeat penalty = disabled or 1.0 | repeat penalty = disabled or 1.0 |
50
+
51
+ * For general use-case: `--temp 1.0 --top-p 0.95`
52
+ * For tool-calling: `--temp 0.7 --top-p 1.0`
53
+ * If using llama.cpp, set `--min-p 0.01` as llama.cpp's default is 0.05
54
+ * Sometimes you'll need to experiment what numbers work best for your use-case.
55
+
56
+ If you experience any issues with these parameters, some users have reported better results when lowering temperature to 0.5-0.6
57
+
58
+ ---
59
+
60
+ ## Benchmarks
61
+
62
+ ![alt="Results Bar Chart"](results_bar_chart.png)
63
+
64
+ ### Model Comparison vs Base
65
+
66
+ ![alt="Comparsion vs Base Chart"](comparison_vs_base.png)
67
+
68
+ - Base model: zai-org/GLM-4.7-Flash
69
+
70
+ | Benchmark | Base Score | Distilled Score | Delta | Delta % |
71
+ |:----------------------|-------------:|--------------:|-------------:|------------:|
72
+ | arc_challenge | **0.224403** | 0.217577 | -0.00682594 | -0.0304183 |
73
+ | gpqa_diamond_zeroshot | 0.262626 | **0.292929** | 0.030303 | 0.115385 |
74
+ | hellaswag | **0.257817** | 0.256722 | -0.0010954 | -0.00424874 |
75
+ | ifeval | 0.109057 | **0.112754** | 0.00369686 | 0.0338983 |
76
+ | mmlu | 0.229454 | **0.240706** | 0.011252 | 0.0490379 |
77
+ | truthfulqa_mc2 | **0.467552** | 0.466805 | -0.000747457 | -0.00159866 |
78
+ | winogrande | 0.468824 | **0.504341** | 0.035517 | 0.0757576 |
79
+
80
+ ### Aggregate Comparison
81
+
82
+ | Benchmarks Compared | Wins vs Base | Ties vs Base | Losses vs Base | Avg Delta |
83
+ |----------------------:|---------------:|---------------:|-----------------:|------------:|
84
+ | 7 | 4 | 0 | 3 | 0.0103 |
85
+
86
+ ### Detailed Results
87
+
88
+ | Model | Benchmark | Score | Total Questions | Total Correct |
89
+ |:-------------------------------------------------------------|:----------------------|---------:|------------------:|----------------:|
90
+ | zai-org/GLM-4.7-Flash | winogrande | 0.468824 | 1267 | 594 |
91
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | winogrande | 0.504341 | 1267 | 639 |
92
+ | zai-org/GLM-4.7-Flash | arc_challenge | 0.224403 | 1172 | 263 |
93
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | arc_challenge | 0.217577 | 1172 | 255 |
94
+ | zai-org/GLM-4.7-Flash | hellaswag | 0.257817 | 10042 | 2589 |
95
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | hellaswag | 0.256722 | 10042 | 2578 |
96
+ | zai-org/GLM-4.7-Flash | truthfulqa_mc2 | 0.467552 | 817 | 381 |
97
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | truthfulqa_mc2 | 0.466805 | 817 | 381 |
98
+ | zai-org/GLM-4.7-Flash | mmlu | 0.229454 | 14042 | 3222 |
99
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | mmlu | 0.240706 | 14042 | 3380 |
100
+ | zai-org/GLM-4.7-Flash | ifeval | 0.109057 | 541 | 59 |
101
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | ifeval | 0.112754 | 541 | 61 |
102
+ | zai-org/GLM-4.7-Flash | gpqa_diamond_zeroshot | 0.262626 | 198 | 52 |
103
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | gpqa_diamond_zeroshot | 0.292929 | 198 | 58 |
104
+
105
+
106
+ ### MMLU Subject Breakdown
107
+
108
+ ![alt="MMLU Subject Breakdown"](mmlu_subject_breakdown.png)
109
+
110
+ #### MMLU Detailed Results
111
+
112
+ | Model | Subject | Benchmark | Score | Total Questions | Total Correct |
113
+ |:-------------------------------------------------------------|:------------------------------------|:-----------------------------------------|---------:|------------------:|----------------:|
114
+ | zai-org/GLM-4.7-Flash | formal_logic | mmlu_formal_logic | 0.285714 | 126 | 36 |
115
+ | zai-org/GLM-4.7-Flash | high_school_european_history | mmlu_high_school_european_history | 0.218182 | 165 | 36 |
116
+ | zai-org/GLM-4.7-Flash | high_school_us_history | mmlu_high_school_us_history | 0.25 | 204 | 51 |
117
+ | zai-org/GLM-4.7-Flash | high_school_world_history | mmlu_high_school_world_history | 0.270042 | 237 | 63 |
118
+ | zai-org/GLM-4.7-Flash | international_law | mmlu_international_law | 0.239669 | 121 | 29 |
119
+ | zai-org/GLM-4.7-Flash | jurisprudence | mmlu_jurisprudence | 0.259259 | 108 | 28 |
120
+ | zai-org/GLM-4.7-Flash | logical_fallacies | mmlu_logical_fallacies | 0.220859 | 163 | 36 |
121
+ | zai-org/GLM-4.7-Flash | moral_disputes | mmlu_moral_disputes | 0.248555 | 346 | 86 |
122
+ | zai-org/GLM-4.7-Flash | moral_scenarios | mmlu_moral_scenarios | 0.237989 | 895 | 213 |
123
+ | zai-org/GLM-4.7-Flash | philosophy | mmlu_philosophy | 0.186495 | 311 | 58 |
124
+ | zai-org/GLM-4.7-Flash | prehistory | mmlu_prehistory | 0.216049 | 324 | 70 |
125
+ | zai-org/GLM-4.7-Flash | professional_law | mmlu_professional_law | 0.245763 | 1534 | 377 |
126
+ | zai-org/GLM-4.7-Flash | world_religions | mmlu_world_religions | 0.321637 | 171 | 55 |
127
+ | zai-org/GLM-4.7-Flash | business_ethics | mmlu_business_ethics | 0.3 | 100 | 30 |
128
+ | zai-org/GLM-4.7-Flash | clinical_knowledge | mmlu_clinical_knowledge | 0.215094 | 265 | 57 |
129
+ | zai-org/GLM-4.7-Flash | college_medicine | mmlu_college_medicine | 0.208092 | 173 | 36 |
130
+ | zai-org/GLM-4.7-Flash | global_facts | mmlu_global_facts | 0.18 | 100 | 18 |
131
+ | zai-org/GLM-4.7-Flash | human_aging | mmlu_human_aging | 0.313901 | 223 | 70 |
132
+ | zai-org/GLM-4.7-Flash | management | mmlu_management | 0.174757 | 103 | 18 |
133
+ | zai-org/GLM-4.7-Flash | marketing | mmlu_marketing | 0.290598 | 234 | 68 |
134
+ | zai-org/GLM-4.7-Flash | medical_genetics | mmlu_medical_genetics | 0.3 | 100 | 30 |
135
+ | zai-org/GLM-4.7-Flash | miscellaneous | mmlu_miscellaneous | 0.237548 | 783 | 186 |
136
+ | zai-org/GLM-4.7-Flash | nutrition | mmlu_nutrition | 0.22549 | 306 | 69 |
137
+ | zai-org/GLM-4.7-Flash | professional_accounting | mmlu_professional_accounting | 0.234043 | 282 | 66 |
138
+ | zai-org/GLM-4.7-Flash | professional_medicine | mmlu_professional_medicine | 0.183824 | 272 | 50 |
139
+ | zai-org/GLM-4.7-Flash | virology | mmlu_virology | 0.283133 | 166 | 47 |
140
+ | zai-org/GLM-4.7-Flash | econometrics | mmlu_econometrics | 0.236842 | 114 | 27 |
141
+ | zai-org/GLM-4.7-Flash | high_school_geography | mmlu_high_school_geography | 0.176768 | 198 | 35 |
142
+ | zai-org/GLM-4.7-Flash | high_school_government_and_politics | mmlu_high_school_government_and_politics | 0.196891 | 193 | 38 |
143
+ | zai-org/GLM-4.7-Flash | high_school_macroeconomics | mmlu_high_school_macroeconomics | 0.202564 | 390 | 79 |
144
+ | zai-org/GLM-4.7-Flash | high_school_microeconomics | mmlu_high_school_microeconomics | 0.214286 | 238 | 51 |
145
+ | zai-org/GLM-4.7-Flash | high_school_psychology | mmlu_high_school_psychology | 0.192661 | 545 | 105 |
146
+ | zai-org/GLM-4.7-Flash | human_sexuality | mmlu_human_sexuality | 0.259542 | 131 | 34 |
147
+ | zai-org/GLM-4.7-Flash | professional_psychology | mmlu_professional_psychology | 0.25 | 612 | 153 |
148
+ | zai-org/GLM-4.7-Flash | public_relations | mmlu_public_relations | 0.218182 | 110 | 24 |
149
+ | zai-org/GLM-4.7-Flash | security_studies | mmlu_security_studies | 0.187755 | 245 | 46 |
150
+ | zai-org/GLM-4.7-Flash | sociology | mmlu_sociology | 0.238806 | 201 | 48 |
151
+ | zai-org/GLM-4.7-Flash | us_foreign_policy | mmlu_us_foreign_policy | 0.28 | 100 | 28 |
152
+ | zai-org/GLM-4.7-Flash | abstract_algebra | mmlu_abstract_algebra | 0.22 | 100 | 22 |
153
+ | zai-org/GLM-4.7-Flash | anatomy | mmlu_anatomy | 0.185185 | 135 | 25 |
154
+ | zai-org/GLM-4.7-Flash | astronomy | mmlu_astronomy | 0.177632 | 152 | 27 |
155
+ | zai-org/GLM-4.7-Flash | college_biology | mmlu_college_biology | 0.256944 | 144 | 37 |
156
+ | zai-org/GLM-4.7-Flash | college_chemistry | mmlu_college_chemistry | 0.2 | 100 | 20 |
157
+ | zai-org/GLM-4.7-Flash | college_computer_science | mmlu_college_computer_science | 0.26 | 100 | 26 |
158
+ | zai-org/GLM-4.7-Flash | college_mathematics | mmlu_college_mathematics | 0.21 | 100 | 21 |
159
+ | zai-org/GLM-4.7-Flash | college_physics | mmlu_college_physics | 0.215686 | 102 | 22 |
160
+ | zai-org/GLM-4.7-Flash | computer_security | mmlu_computer_security | 0.28 | 100 | 28 |
161
+ | zai-org/GLM-4.7-Flash | conceptual_physics | mmlu_conceptual_physics | 0.26383 | 235 | 62 |
162
+ | zai-org/GLM-4.7-Flash | electrical_engineering | mmlu_electrical_engineering | 0.241379 | 145 | 35 |
163
+ | zai-org/GLM-4.7-Flash | elementary_mathematics | mmlu_elementary_mathematics | 0.208995 | 378 | 79 |
164
+ | zai-org/GLM-4.7-Flash | high_school_biology | mmlu_high_school_biology | 0.174194 | 310 | 54 |
165
+ | zai-org/GLM-4.7-Flash | high_school_chemistry | mmlu_high_school_chemistry | 0.152709 | 203 | 31 |
166
+ | zai-org/GLM-4.7-Flash | high_school_computer_science | mmlu_high_school_computer_science | 0.25 | 100 | 25 |
167
+ | zai-org/GLM-4.7-Flash | high_school_mathematics | mmlu_high_school_mathematics | 0.211111 | 270 | 57 |
168
+ | zai-org/GLM-4.7-Flash | high_school_physics | mmlu_high_school_physics | 0.198675 | 151 | 29 |
169
+ | zai-org/GLM-4.7-Flash | high_school_statistics | mmlu_high_school_statistics | 0.152778 | 216 | 33 |
170
+ | zai-org/GLM-4.7-Flash | machine_learning | mmlu_machine_learning | 0.321429 | 112 | 36 |
171
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | formal_logic | mmlu_formal_logic | 0.206349 | 126 | 26 |
172
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_european_history | mmlu_high_school_european_history | 0.206061 | 165 | 34 |
173
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_us_history | mmlu_high_school_us_history | 0.245098 | 204 | 50 |
174
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_world_history | mmlu_high_school_world_history | 0.270042 | 237 | 63 |
175
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | international_law | mmlu_international_law | 0.239669 | 121 | 29 |
176
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | jurisprudence | mmlu_jurisprudence | 0.305556 | 108 | 33 |
177
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | logical_fallacies | mmlu_logical_fallacies | 0.214724 | 163 | 35 |
178
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | moral_disputes | mmlu_moral_disputes | 0.271676 | 346 | 93 |
179
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | moral_scenarios | mmlu_moral_scenarios | 0.222346 | 895 | 199 |
180
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | philosophy | mmlu_philosophy | 0.228296 | 311 | 71 |
181
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | prehistory | mmlu_prehistory | 0.271605 | 324 | 88 |
182
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | professional_law | mmlu_professional_law | 0.252934 | 1534 | 388 |
183
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | world_religions | mmlu_world_religions | 0.280702 | 171 | 48 |
184
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | business_ethics | mmlu_business_ethics | 0.3 | 100 | 30 |
185
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | clinical_knowledge | mmlu_clinical_knowledge | 0.267925 | 265 | 71 |
186
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_medicine | mmlu_college_medicine | 0.213873 | 173 | 37 |
187
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | global_facts | mmlu_global_facts | 0.32 | 100 | 32 |
188
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | human_aging | mmlu_human_aging | 0.327354 | 223 | 73 |
189
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | management | mmlu_management | 0.213592 | 103 | 22 |
190
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | marketing | mmlu_marketing | 0.286325 | 234 | 67 |
191
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | medical_genetics | mmlu_medical_genetics | 0.35 | 100 | 35 |
192
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | miscellaneous | mmlu_miscellaneous | 0.254151 | 783 | 199 |
193
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | nutrition | mmlu_nutrition | 0.222222 | 306 | 68 |
194
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | professional_accounting | mmlu_professional_accounting | 0.244681 | 282 | 69 |
195
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | professional_medicine | mmlu_professional_medicine | 0.183824 | 272 | 50 |
196
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | virology | mmlu_virology | 0.325301 | 166 | 54 |
197
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | econometrics | mmlu_econometrics | 0.280702 | 114 | 32 |
198
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_geography | mmlu_high_school_geography | 0.207071 | 198 | 41 |
199
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_government_and_politics | mmlu_high_school_government_and_politics | 0.176166 | 193 | 34 |
200
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_macroeconomics | mmlu_high_school_macroeconomics | 0.217949 | 390 | 85 |
201
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_microeconomics | mmlu_high_school_microeconomics | 0.222689 | 238 | 53 |
202
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_psychology | mmlu_high_school_psychology | 0.209174 | 545 | 114 |
203
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | human_sexuality | mmlu_human_sexuality | 0.21374 | 131 | 28 |
204
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | professional_psychology | mmlu_professional_psychology | 0.259804 | 612 | 159 |
205
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | public_relations | mmlu_public_relations | 0.309091 | 110 | 34 |
206
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | security_studies | mmlu_security_studies | 0.159184 | 245 | 39 |
207
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | sociology | mmlu_sociology | 0.253731 | 201 | 51 |
208
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | us_foreign_policy | mmlu_us_foreign_policy | 0.25 | 100 | 25 |
209
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | abstract_algebra | mmlu_abstract_algebra | 0.23 | 100 | 23 |
210
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | anatomy | mmlu_anatomy | 0.251852 | 135 | 34 |
211
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | astronomy | mmlu_astronomy | 0.164474 | 152 | 25 |
212
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_biology | mmlu_college_biology | 0.263889 | 144 | 38 |
213
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_chemistry | mmlu_college_chemistry | 0.22 | 100 | 22 |
214
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_computer_science | mmlu_college_computer_science | 0.22 | 100 | 22 |
215
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_mathematics | mmlu_college_mathematics | 0.25 | 100 | 25 |
216
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | college_physics | mmlu_college_physics | 0.245098 | 102 | 25 |
217
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | computer_security | mmlu_computer_security | 0.24 | 100 | 24 |
218
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | conceptual_physics | mmlu_conceptual_physics | 0.340426 | 235 | 80 |
219
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | electrical_engineering | mmlu_electrical_engineering | 0.193103 | 145 | 28 |
220
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | elementary_mathematics | mmlu_elementary_mathematics | 0.240741 | 378 | 91 |
221
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_biology | mmlu_high_school_biology | 0.190323 | 310 | 58 |
222
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_chemistry | mmlu_high_school_chemistry | 0.216749 | 203 | 44 |
223
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_computer_science | mmlu_high_school_computer_science | 0.19 | 100 | 19 |
224
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_mathematics | mmlu_high_school_mathematics | 0.240741 | 270 | 65 |
225
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_physics | mmlu_high_school_physics | 0.172185 | 151 | 26 |
226
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | high_school_statistics | mmlu_high_school_statistics | 0.194444 | 216 | 42 |
227
+ | TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill | machine_learning | mmlu_machine_learning | 0.241071 | 112 | 27 |
228
+
229
+ ### Benchmark Config
230
+
231
+ - **Quantization:** 4bit
232
+ - **Temperature:** 0.0
233
+ - **Top P:** 1.0
234
+ - **Top K:** 0
235
+ - **Repetition Penalty:** 1.0
236
+
237
+ All results were obtained through the official lm evaluation harness
238
+
239
+ ---
240
+ This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
comparison_vs_base.png ADDED
glm-4.7-flash-claude-4.5-opus.bf16.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a331855f92815e4c80ace5bdaf2978f2ece2f997dbb1d24cbb621cc1bb4553b
3
+ size 59908837568
glm-4.7-flash-claude-4.5-opus.f16.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5fbdfaf8fce25fcb1f04c7dbeb25bcb1ebd68d34450cae154b2d2ef94ef5b33
3
+ size 59908837568
glm-4.7-flash-claude-4.5-opus.iq2_m.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4088e83d8988a861569fcede22d1e6f0c92f05c766ae0196e07c9e91eb3d5c9
3
+ size 9898321504
glm-4.7-flash-claude-4.5-opus.iq3_m.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4a20371d6aafba6c114cfa693e061394b155f41506d191b705afafbfabd6faa
3
+ size 13211266656
glm-4.7-flash-claude-4.5-opus.iq3_xs.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ad3b52cdd1f9fdd6a19ae5c24a3641bca1e91318600b38f4a9701e4cce230e3
3
+ size 12344307296
glm-4.7-flash-claude-4.5-opus.iq4_nl.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2eaa3f415c6563498589cf80241498e13fad82c72a5731fd9cce1c309b95d05a
3
+ size 16956023392
glm-4.7-flash-claude-4.5-opus.iq4_xs.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59e01173e6ec7238f6b234706fd3cc0a78cca33b43337cef3af140e72a922446
3
+ size 16033288800
glm-4.7-flash-claude-4.5-opus.q3_k_m.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30b9765dde07c4b27744917516d6e521f8e06fb3199a1b265e77e938239fd48b
3
+ size 14380018880
glm-4.7-flash-claude-4.5-opus.q3_k_s.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea66008c8398a90a856fcfc277a22f8b17fbc6f67308e518428216dc265ceb95
3
+ size 13034400960
glm-4.7-flash-claude-4.5-opus.q4_k_m.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b7f1a37ddac09fc3a5cc4ff9a78410c30cd4a105edb4ee6b87194c9523b1c88
3
+ size 18132721856
glm-4.7-flash-claude-4.5-opus.q5_k_m.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4023664a7445a2ff77f29f644045ab095ecdf26fee1681e9f3b6eeb7c1332ed2
3
+ size 21264515264
glm-4.7-flash-claude-4.5-opus.q6_k.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:133901b9d3de9e3b0ea12bc5f9085d7d0aef275c83671bd7df187e0a11eb45c2
3
+ size 24614786240
glm-4.7-flash-claude-4.5-opus.q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c857f56b727287f28e3869f08b1db37e08546992e67944e3d5d39b418c0ae367
3
+ size 31842799808
mmlu_subject_breakdown.png ADDED

Git LFS Details

  • SHA256: b3f7b86a080c34460f4adb41dc876559f4d3ba3e7ba06524829cc6a66e3b2cb2
  • Pointer size: 132 Bytes
  • Size of remote file: 1.28 MB
results_bar_chart.png ADDED

Git LFS Details

  • SHA256: 42906c3a6d6514111e5c0049d23e991c51d9c523b06a5a3e8f94cffe1e6455ee
  • Pointer size: 131 Bytes
  • Size of remote file: 171 kB