Hugging Face's logo

inclusionAI
/
Ling-1T

Text Generation
Transformers
Safetensors
bailing_moe
conversational
custom_code
Model card Files
xet
Community
17

Update README.md

#1
by m1ngcheng - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+51
-55
Files changed (1) hide show
  1. README.md +51 -55
README.md CHANGED
@@ -4,13 +4,11 @@ pipeline_tag: text-generation
4
  library_name: transformers
5
  ---
6
 
7
-
8
-
9
  <p align="center">
10
  <img src="https://mdn.alipayobjects.com/huamei_qa8qxu/afts/img/A*4QxcQrBlTiAAAAAAQXAAAAgAemJ7AQ/original" width="100"/>
11
- <p>
12
 
13
- <p align="center">πŸ€— <a href="https://huggingface.co/inclusionAI">Hugging Face</a>&nbsp&nbsp | &nbsp&nbspπŸ€– <a href="https://modelscope.cn/organization/inclusionAI">ModelScope </a>&nbsp&nbsp | &nbsp&nbspπŸ™ <a href="https://zenmux.ai/inclusionai/ling-1t?utm_source=hf_inclusionAI">Experience Now</a></p>
14
 
15
 
16
  ## Introduction
@@ -89,49 +87,49 @@ Ling-1T has been extensively evaluated across **knowledge**, **code**, **math**,
89
  It currently stands as the **best open-source flagship non-thinking model**, rivaling closed-source APIs in complex reasoning while maintaining exceptional efficiency and interpretability.
90
 
91
  ## Evaluation
92
- | Task | Benchmark | DeepSeek-V3.1-Teminus | Kimi-K2-Instruct-0905 | gpt-5-main | Gemini 2.5 Pro | Ling-1T |
93
- |------|------------|--------------------------------------|------------------------|-------------|-----------------------------|----------|
94
- | | | (NonThinking) | | | (thinkBudget=128) | |
95
- | **Knowledge** | **Professional Knowledge** | | | | | |
96
- | | C-Eval | __91.76__ | 91.12 | 83.59 | 88.77 | __<span style="color:red">92.19</span>__ |
97
- | | MMLU-Redux (EM) | 92.37 | 91.58 | **92.75** | __<span style="color:red">94.67</span>__ | 92.25 |
98
- | | MMLU-Pro | __<span style="color:red">83.25</span>__ | 81.03 | 81.94 | **82.13** | 82.04 |
99
- | **Knowledge** | **STEM** | | | | | |
100
- | | MMLU-Pro-Stem | 87.91 | 85.30 | 73.45 | __<span style="color:red">88.60</span>__ | **88.5** |
101
- | | OlympiadBench-stem | 87.83 | 79.13 | 78.26 | **89.57** | __<span style="color:red">91.3</span>__ |
102
- | | GPQA-Diamond | __<span style="color:red">76.23</span>__ | **73.93** | 71.31 | 71.81 | 72.98 |
103
- | **Coding** | **Code Generation** | | | | | |
104
- | | MultiPL-E | **77.68** | 73.76 | 76.66 | 71.48 | __<span style="color:red">77.91</span>__ |
105
- | | mbpp | 90.69 | 89.96 | **91.72** | 91.01 | __<span style="color:red">96.87</span>__ |
106
- | | LiveCodeBench (2408-2505) | 48.02 | 48.95 | **48.57** | 45.43 | __<span style="color:red">61.68</span>__ |
107
- | | CodeForces-rating | 1582 | 1574 | 1120 | **1675** | __<span style="color:red">1901</span>__ |
108
- | | BIRD_SQL | 44.88 | 46.45 | 43.97 | __<span style="color:red">54.76</span>__ | **52.38** |
109
- | **Coding** | **Software Development** | | | | | |
110
- | | ArtifactsBench | 43.29 | 44.87 | 41.04 | __<span style="color:red">60.28</span>__ | **59.31** |
111
- | | FullStack Bench | **55.48** | 54.00 | 50.92 | 48.19 | __<span style="color:red">56.55</span>__ |
112
- | | Aider | **88.16** | 85.34 | 84.40 | __<span style="color:red">89.85</span>__ | 83.65 |
113
- | **Math** | **Competition Math** | | | | | |
114
- | | CNMO 2024 | 73.78 | 68.92 | 63.11 | **74.65** | __<span style="color:red">79.25</span>__ |
115
- | | AIME 2025 | 55.21 | 50.16 | 59.43 | **70.10** | __<span style="color:red">70.42</span>__ |
116
- | | UGMathBench | **72.70** | 69.97 | 67.27 | 70.10 | __<span style="color:red">74.95</span>__ |
117
- | | Omni-Math | 64.77 | 62.42 | 61.09 | **72.02** | __<span style="color:red">74.46</span>__ |
118
- | **Math** | **Professional Math** | | | | | |
119
- | | FinanceReasoning | 86.44 | 84.83 | 86.28 | **86.65** | __<span style="color:red">87.45</span>__ |
120
- | | Optibench | 64.30 | 60.83 | 40.06 | **68.76** | __<span style="color:red">74.71</span>__ |
121
- | | OptMATH | 35.99 | 35.84 | 39.16 | **42.77** | __<span style="color:red">57.68</span>__ |
122
- | **General Reasoning** | | | | | | |
123
- | | BBEH | **42.86** | 34.83 | 39.75 | 29.08 | __<span style="color:red">47.34</span>__ |
124
- | | KOR-Bench | **73.76** | 73.20 | 70.56 | 59.68 | __<span style="color:red">76.00</span>__ |
125
- | | ARC-AGI-1 | 14.69 | **22.19** | 14.06 | 18.94 | __<span style="color:red">43.81</span>__ |
126
- | | ZebraLogic | 81.6 | **85.5** | 57.3 | 70.2 | __<span style="color:red">90.8</span>__ |
127
- | **Agent** | | | | | | |
128
- | | BFCL-V3 | 52.67 | __<span style="color:red">71.05</span>__ | 50.27 | 63.31 | **69.64** |
129
- | **Alignment** | | | | | | |
130
- | | Arena Hard V2 ELO | 54.09 | __<span style="color:red">76.95</span>__ | 68.37 | 65.37 | **76.26** |
131
- | | Arena Hard V2 Win Rate | 63.24 | 69.88 | 65.06 | **74.46** | __<span style="color:red">75.83</span>__ |
132
- | | writing_bench | 80.95 | **87.59** | 77.07 | 80.53 | __<span style="color:red">89.4</span>__ |
133
- | | Creative Writing v3 | 85.18 | **87.01** | 80.93 | 84.99 | <span style="color:red">89.24</span> |
134
- | | MultiChallenge | 42.49 | 48.72 | 48.72 | **51.28** | __<span style="color:red">58.24</span>__ |
135
 
136
 
137
 
@@ -141,9 +139,9 @@ You can download Ling-1T from the following table. If you are located in mainlan
141
 
142
  <center>
143
 
144
- | **Model** | **Context Length** | **Download** |
145
- |:----------------------:| :----------------: |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
146
- | Ling-1T | 32K -> 128K (YaRN) | [πŸ€— HuggingFace](https://huggingface.co/inclusionAI/Ling-1T) &nbsp;&nbsp; [πŸ€– ModelScope](https://www.modelscope.cn/models/inclusionAI/Ling-1T) |
147
 
148
  </center>
149
 
@@ -228,7 +226,7 @@ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
228
 
229
  ### πŸ€– ModelScope
230
 
231
- If you're in mainland China, we strongly recommend you to use our model from πŸ€– <a href="https://modelscope.cn/organization/inclusionAI">ModelScope</a>.
232
 
233
  ## Deployment
234
 
@@ -322,7 +320,7 @@ BF16 and FP8 models are supported by SGLang now, it depends on the dtype of the
322
  - Start server:
323
  ```bash
324
  python -m sglang.launch_server \
325
- --model-path $MODLE_PATH \
326
  --host 0.0.0.0 --port $PORT \
327
  --trust-remote-code \
328
  --attention-backend fa3
@@ -355,8 +353,6 @@ While **Ling-1T** has made strong progress in efficient reasoning, cross-domain
355
  Ling-1T will continue to evolve in architecture, reasoning, and alignment, advancing the series toward more general intelligence.
356
 
357
 
358
-
359
  ## License
360
 
361
- This code repository is licensed under [the MIT License](https://github.com/inclusionAI/Ling-V2/blob/main/LICENSE).
362
-
 
4
  library_name: transformers
5
  ---
6
 
 
 
7
  <p align="center">
8
  <img src="https://mdn.alipayobjects.com/huamei_qa8qxu/afts/img/A*4QxcQrBlTiAAAAAAQXAAAAgAemJ7AQ/original" width="100"/>
9
+ </p>
10
 
11
+ <p align="center">πŸ€— <a href="https://huggingface.co/inclusionAI">Hugging Face</a>&nbsp;&nbsp; | &nbsp;&nbsp;πŸ€– <a href="https://modelscope.cn/organization/inclusionAI">ModelScope </a>&nbsp;&nbsp; | &nbsp;&nbsp;πŸ™ <a href="https://zenmux.ai/inclusionai/ling-1t?utm_source=hf_inclusionAI">Experience Now</a></p>
12
 
13
 
14
  ## Introduction
 
87
  It currently stands as the **best open-source flagship non-thinking model**, rivaling closed-source APIs in complex reasoning while maintaining exceptional efficiency and interpretability.
88
 
89
  ## Evaluation
90
+ | Task | Benchmark | DeepSeek-V3.1-Terminus | Kimi-K2-Instruct-0905 | gpt-5-main | Gemini 2.5 Pro | Ling-1T |
91
+ | --------------------- | -------------------------- | ---------------------------------------- | ---------------------------------------- | ---------- | ---------------------------------------- | ---------------------------------------- |
92
+ | | | (NonThinking) | | | (thinkBudget=128) | |
93
+ | **Knowledge** | **Professional Knowledge** | | | | | |
94
+ | | C-Eval | __91.76__ | 91.12 | 83.59 | 88.77 | __<span style="color:red">92.19</span>__ |
95
+ | | MMLU-Redux (EM) | 92.37 | 91.58 | **92.75** | __<span style="color:red">94.67</span>__ | 92.25 |
96
+ | | MMLU-Pro | __<span style="color:red">83.25</span>__ | 81.03 | 81.94 | **82.13** | 82.04 |
97
+ | **Knowledge** | **STEM** | | | | | |
98
+ | | MMLU-Pro-Stem | 87.91 | 85.30 | 73.45 | __<span style="color:red">88.60</span>__ | **88.5** |
99
+ | | OlympiadBench-stem | 87.83 | 79.13 | 78.26 | **89.57** | __<span style="color:red">91.3</span>__ |
100
+ | | GPQA-Diamond | __<span style="color:red">76.23</span>__ | **73.93** | 71.31 | 71.81 | 72.98 |
101
+ | **Coding** | **Code Generation** | | | | | |
102
+ | | MultiPL-E | **77.68** | 73.76 | 76.66 | 71.48 | __<span style="color:red">77.91</span>__ |
103
+ | | mbpp | 90.69 | 89.96 | **91.72** | 91.01 | __<span style="color:red">96.87</span>__ |
104
+ | | LiveCodeBench (2408-2505) | 48.02 | 48.95 | **48.57** | 45.43 | __<span style="color:red">61.68</span>__ |
105
+ | | CodeForces-rating | 1582 | 1574 | 1120 | **1675** | __<span style="color:red">1901</span>__ |
106
+ | | BIRD_SQL | 44.88 | 46.45 | 43.97 | __<span style="color:red">54.76</span>__ | **52.38** |
107
+ | **Coding** | **Software Development** | | | | | |
108
+ | | ArtifactsBench | 43.29 | 44.87 | 41.04 | __<span style="color:red">60.28</span>__ | **59.31** |
109
+ | | FullStack Bench | **55.48** | 54.00 | 50.92 | 48.19 | __<span style="color:red">56.55</span>__ |
110
+ | | Aider | **88.16** | 85.34 | 84.40 | __<span style="color:red">89.85</span>__ | 83.65 |
111
+ | **Math** | **Competition Math** | | | | | |
112
+ | | CNMO 2024 | 73.78 | 68.92 | 63.11 | **74.65** | __<span style="color:red">79.25</span>__ |
113
+ | | AIME 2025 | 55.21 | 50.16 | 59.43 | **70.10** | __<span style="color:red">70.42</span>__ |
114
+ | | UGMathBench | **72.70** | 69.97 | 67.27 | 70.10 | __<span style="color:red">74.95</span>__ |
115
+ | | Omni-Math | 64.77 | 62.42 | 61.09 | **72.02** | __<span style="color:red">74.46</span>__ |
116
+ | **Math** | **Professional Math** | | | | | |
117
+ | | FinanceReasoning | 86.44 | 84.83 | 86.28 | **86.65** | __<span style="color:red">87.45</span>__ |
118
+ | | Optibench | 64.30 | 60.83 | 40.06 | **68.76** | __<span style="color:red">74.71</span>__ |
119
+ | | OptMATH | 35.99 | 35.84 | 39.16 | **42.77** | __<span style="color:red">57.68</span>__ |
120
+ | **General Reasoning** | | | | | | |
121
+ | | BBEH | **42.86** | 34.83 | 39.75 | 29.08 | __<span style="color:red">47.34</span>__ |
122
+ | | KOR-Bench | **73.76** | 73.20 | 70.56 | 59.68 | __<span style="color:red">76.00</span>__ |
123
+ | | ARC-AGI-1 | 14.69 | **22.19** | 14.06 | 18.94 | __<span style="color:red">43.81</span>__ |
124
+ | | ZebraLogic | 81.6 | **85.5** | 57.3 | 70.2 | __<span style="color:red">90.8</span>__ |
125
+ | **Agent** | | | | | | |
126
+ | | BFCL-V3 | 52.67 | __<span style="color:red">71.05</span>__ | 50.27 | 63.31 | **69.64** |
127
+ | **Alignment** | | | | | | |
128
+ | | Arena Hard V2 ELO | 54.09 | __<span style="color:red">76.95</span>__ | 68.37 | 65.37 | **76.26** |
129
+ | | Arena Hard V2 Win Rate | 63.24 | 69.88 | 65.06 | **74.46** | __<span style="color:red">75.83</span>__ |
130
+ | | writing_bench | 80.95 | **87.59** | 77.07 | 80.53 | __<span style="color:red">89.4</span>__ |
131
+ | | Creative Writing v3 | 85.18 | **87.01** | 80.93 | 84.99 | <span style="color:red">89.24</span> |
132
+ | | MultiChallenge | 42.49 | 48.72 | 48.72 | **51.28** | __<span style="color:red">58.24</span>__ |
133
 
134
 
135
 
 
139
 
140
  <center>
141
 
142
+ | **Model** | **Context Length** | **Download** |
143
+ | :-------: | :----------------: | :-------------------------------------------------------------------------------------------------------------------------------------------: |
144
+ | Ling-1T | 32K -> 128K (YaRN) | [πŸ€— HuggingFace](https://huggingface.co/inclusionAI/Ling-1T) &nbsp;&nbsp; [πŸ€– ModelScope](https://www.modelscope.cn/models/inclusionAI/Ling-1T) |
145
 
146
  </center>
147
 
 
226
 
227
  ### πŸ€– ModelScope
228
 
229
+ If you're in mainland China, we strongly recommend you to use our model from πŸ€– <a href="https://modelscope.cn/models/inclusionAI/Ling-1T">ModelScope</a>.
230
 
231
  ## Deployment
232
 
 
320
  - Start server:
321
  ```bash
322
  python -m sglang.launch_server \
323
+ --model-path $MODEL_PATH \
324
  --host 0.0.0.0 --port $PORT \
325
  --trust-remote-code \
326
  --attention-backend fa3
 
353
  Ling-1T will continue to evolve in architecture, reasoning, and alignment, advancing the series toward more general intelligence.
354
 
355
 
 
356
  ## License
357
 
358
+ This code repository is licensed under [the MIT License](https://github.com/inclusionAI/Ling-V2/blob/main/LICENSE).