| | --- |
| | license: mit |
| | pipeline_tag: text-generation |
| | library_name: transformers |
| | --- |
| | |
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_qa8qxu/afts/img/A*4QxcQrBlTiAAAAAAQXAAAAgAemJ7AQ/original" width="100"/> |
| | </p> |
| | |
| | <p align="center">🤗 <a href="https://huggingface.co/inclusionAI">Hugging Face</a> | 🤖 <a href="https://modelscope.cn/organization/inclusionAI">ModelScope </a> | 🐙 <a href="https://zenmux.ai/inclusionai/ling-1t?utm_source=hf_inclusionAI">Experience Now</a></p> |
| |
|
| |
|
| | ## Introduction |
| |
|
| | **Ling-1T** is the first flagship *non-thinking* model in the Ling 2.0 series, featuring **1 trillion total parameters** with **≈ 50 billion active parameters per token**. |
| | Built on the Ling 2.0 architecture, Ling-1T is designed to push the limits of *efficient reasoning* and *scalable cognition*. |
| |
|
| | Pre-trained on **20 trillion+ high-quality, reasoning-dense tokens**, Ling-1T-base supports up to **128K context length** and adopts an **evolutionary chain-of-thought (Evo-CoT)** process across mid-training and post-training. |
| | This curriculum greatly enhances the model’s efficiency and reasoning depth, allowing Ling-1T to achieve **state-of-the-art performance** on multiple complex reasoning benchmarks—balancing **accuracy** and **efficiency**. |
| |
|
| |
|
| | ### Flagship-Level Efficient Reasoning |
| |
|
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/FRNXSJFZGXkAAAAAT-AAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/3in4SJr8YPkAAAAAUNAAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | We comprehensively evaluated Ling-1T against leading flagship models, including both **open-source giants** (e.g., *DeepSeek-V3.1-Terminus*, *Kimi-K2-Instruct-0905*) and **closed-source APIs** (*GPT-5-main*, *Gemini-2.5-Pro*). |
| | Across code generation, software development, competition-level mathematics, professional math, and logical reasoning, Ling-1T consistently demonstrates **superior complex reasoning ability** and overall advantage. |
| |
|
| | In the **AIME 25** benchmark, Ling-1T extends the **Pareto frontier** of reasoning accuracy vs. reasoning length, showcasing its strength in **“efficient thinking and precise reasoning.”** |
| |
|
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/J8ciS5KbIrwAAAAAceAAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | ### Aesthetic Understanding and Front-End Generation |
| |
|
| | Ling-1T excels in visual reasoning and front-end code generation tasks, combining deep semantic understanding with precise code synthesis. |
| | We introduce a hybrid *Syntax–Function–Aesthetics* reward mechanism, enabling the model to not only generate correct and functional code but also demonstrate a refined sense of **visual aesthetics**. |
| | On **ArtifactsBench**, Ling-1T ranks **first among open-source models**, and the benchmark visualizations in this card were, in fact, *generated by Ling-1T itself*. |
| |
|
| |
|
| | ### Emergent Intelligence at Trillion-Scale |
| |
|
| | Scaling to the trillion-parameter level has revealed strong **emergent reasoning and transfer capabilities**. |
| | For example, in the **BFCL V3** tool-use benchmark, Ling-1T achieves **≈ 70% tool-call accuracy** with only light instruction tuning—despite having seen no large-scale trajectory data during training. |
| | Ling-1T can: |
| |
|
| | * Interpret complex natural-language instructions |
| | * Transform abstract logic into functional visual components |
| | * Generate cross-platform compatible front-end code |
| | * Create stylistically controlled marketing copy and multi-lingual text |
| |
|
| | These capabilities form the foundation for **general, collaborative human–AI intelligence**, which we aim to advance together with the open-source community through Ling-1T’s release. |
| |
|
| |
|
| | ### Pre-Training at Trillion Scale |
| |
|
| | The Ling 2.0 architecture was designed from the ground up for trillion-scale efficiency, guided by the **Ling Scaling Law** ([arXiv:2507.17702](https://arxiv.org/abs/2507.17702)). |
| | This ensures architectural and hyperparameter scalability even under **1e25–1e26 FLOPs** of compute. |
| |
|
| | Key architectural innovations include: |
| |
|
| | * **1T total / 50B active parameters** with a **1/32 MoE activation ratio** |
| | * **MTP layers** for enhanced compositional reasoning |
| | * **Aux-loss-free**, **sigmoid-scoring expert routing** with **zero-mean updates** |
| | * **QK Normalization** for fully stable convergence |
| |
|
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/03WMQZIYxpUAAAAAVTAAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | Ling-1T is the **largest FP8-trained foundation model** known to date. |
| | FP8 mixed-precision training yields **15%+ end-to-end speedup**, improved memory efficiency, and maintains **≤ 0.1% loss deviation** from BF16 across **1T tokens**. |
| | A fine-grained, **heterogeneous 1F1B interleaved pipeline** further boosts utilization by 40 %+. |
| | System-level optimizations—fused kernels, communication scheduling, recomputation, checkpointing, simulation, and telemetry—ensure stable trillion-scale training. |
| |
|
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/y5UVSKACgLEAAAAAVcAAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | Pre-training used over **20T high-quality tokens**, with **> 40% reasoning-dense data** in later stages. |
| | Mid-training introduced **curated chain-of-thought corpora** for “**reasoning pre-activation**”, improving downstream reasoning stability. |
| | A custom **WSM (Warmup–Stable–Merge)** LR scheduler with mid-train checkpoint merging simulates LR decay and boosts generalization. |
| |
|
| |
|
| | ### Post-Training and Evo-CoT Optimization |
| |
|
| | Built upon mid-training reasoning activation, post-training adopts **Evo-CoT (Evolutionary Chain-of-Thought)** for progressive reasoning enhancement under controllable cost. |
| | This approach continually expands the **Pareto frontier** of reasoning accuracy vs. efficiency—ideal for reflexive non-thinking models. |
| |
|
| | For reinforcement learning, we introduce **LPO (Linguistics-Unit Policy Optimization)** —a novel sentence-level policy optimization method. |
| | Unlike GRPO (token-level) or GSPO (sequence-level) algorithms, LPO treats *sentences* as the natural semantic action units, enabling precise alignment between rewards and reasoning behavior. |
| | Empirically, LPO offers superior **training stability** and **generalization** across reasoning tasks. |
| |
|
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/kbEWT4BGEQQAAAAAWwAAAAgADkV7AQFr/original"/> |
| | <p> |
| | <p align="center"> |
| | <img src="https://mdn.alipayobjects.com/huamei_bcz3yt/afts/img/aF5LRqK5LMcAAAAAZHAAAAgADkV7AQFr/original"/> |
| | <p> |
| | |
| | ## Evaluation |
| |
|
| | Ling-1T has been extensively evaluated across **knowledge**, **code**, **math**, **reasoning**, **agent**, and **alignment** benchmarks. |
| | 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. |
| |
|
| | ## Evaluation |
| | | Task | Benchmark | DeepSeek-V3.1-Terminus | Kimi-K2-Instruct-0905 | gpt-5-main | Gemini 2.5 Pro | Ling-1T | |
| | | --------------------- | -------------------------- | ---------------------------------------- | ---------------------------------------- | ---------- | ---------------------------------------- | ---------------------------------------- | |
| | | | | (NonThinking) | | | (thinkBudget=128) | | |
| | | **Knowledge** | **Professional Knowledge** | | | | | | |
| | | | C-Eval | __91.76__ | 91.12 | 83.59 | 88.77 | __<span style="color:red">92.19</span>__ | |
| | | | MMLU-Redux (EM) | 92.37 | 91.58 | **92.75** | __<span style="color:red">94.67</span>__ | 92.25 | |
| | | | MMLU-Pro | __<span style="color:red">83.25</span>__ | 81.03 | 81.94 | **82.13** | 82.04 | |
| | | **Knowledge** | **STEM** | | | | | | |
| | | | MMLU-Pro-Stem | 87.91 | 85.30 | 73.45 | __<span style="color:red">88.60</span>__ | **88.5** | |
| | | | OlympiadBench-stem | 87.83 | 79.13 | 78.26 | **89.57** | __<span style="color:red">91.3</span>__ | |
| | | | GPQA-Diamond | __<span style="color:red">76.23</span>__ | **73.93** | 71.31 | 71.81 | 72.98 | |
| | | **Coding** | **Code Generation** | | | | | | |
| | | | MultiPL-E | **77.68** | 73.76 | 76.66 | 71.48 | __<span style="color:red">77.91</span>__ | |
| | | | mbpp | 90.69 | 89.96 | **91.72** | 91.01 | __<span style="color:red">96.87</span>__ | |
| | | | LiveCodeBench (2408-2505) | 48.02 | 48.95 | **48.57** | 45.43 | __<span style="color:red">61.68</span>__ | |
| | | | CodeForces-rating | 1582 | 1574 | 1120 | **1675** | __<span style="color:red">1901</span>__ | |
| | | | BIRD_SQL | 44.88 | 46.45 | 43.97 | __<span style="color:red">54.76</span>__ | **52.38** | |
| | | **Coding** | **Software Development** | | | | | | |
| | | | ArtifactsBench | 43.29 | 44.87 | 41.04 | __<span style="color:red">60.28</span>__ | **59.31** | |
| | | | FullStack Bench | **55.48** | 54.00 | 50.92 | 48.19 | __<span style="color:red">56.55</span>__ | |
| | | | Aider | **88.16** | 85.34 | 84.40 | __<span style="color:red">89.85</span>__ | 83.65 | |
| | | **Math** | **Competition Math** | | | | | | |
| | | | CNMO 2024 | 73.78 | 68.92 | 63.11 | **74.65** | __<span style="color:red">79.25</span>__ | |
| | | | AIME 2025 | 55.21 | 50.16 | 59.43 | **70.10** | __<span style="color:red">70.42</span>__ | |
| | | | UGMathBench | **72.70** | 69.97 | 67.27 | 70.10 | __<span style="color:red">74.95</span>__ | |
| | | | Omni-Math | 64.77 | 62.42 | 61.09 | **72.02** | __<span style="color:red">74.46</span>__ | |
| | | **Math** | **Professional Math** | | | | | | |
| | | | FinanceReasoning | 86.44 | 84.83 | 86.28 | **86.65** | __<span style="color:red">87.45</span>__ | |
| | | | Optibench | 64.30 | 60.83 | 40.06 | **68.76** | __<span style="color:red">74.71</span>__ | |
| | | | OptMATH | 35.99 | 35.84 | 39.16 | **42.77** | __<span style="color:red">57.68</span>__ | |
| | | **General Reasoning** | | | | | | | |
| | | | BBEH | **42.86** | 34.83 | 39.75 | 29.08 | __<span style="color:red">47.34</span>__ | |
| | | | KOR-Bench | **73.76** | 73.20 | 70.56 | 59.68 | __<span style="color:red">76.00</span>__ | |
| | | | ARC-AGI-1 | 14.69 | **22.19** | 14.06 | 18.94 | __<span style="color:red">43.81</span>__ | |
| | | | ZebraLogic | 81.6 | **85.5** | 57.3 | 70.2 | __<span style="color:red">90.8</span>__ | |
| | | **Agent** | | | | | | | |
| | | | BFCL-V3 | 52.67 | __<span style="color:red">71.05</span>__ | 50.27 | 63.31 | **69.64** | |
| | | **Alignment** | | | | | | | |
| | | | Arena Hard V2 ELO | 54.09 | __<span style="color:red">76.95</span>__ | 68.37 | 65.37 | **76.26** | |
| | | | Arena Hard V2 Win Rate | 63.24 | 69.88 | 65.06 | **74.46** | __<span style="color:red">75.83</span>__ | |
| | | | writing_bench | 80.95 | **87.59** | 77.07 | 80.53 | __<span style="color:red">89.4</span>__ | |
| | | | Creative Writing v3 | 85.18 | **87.01** | 80.93 | 84.99 | <span style="color:red">89.24</span> | |
| | | | MultiChallenge | 42.49 | 48.72 | 48.72 | **51.28** | __<span style="color:red">58.24</span>__ | |
| |
|
| |
|
| |
|
| | ## Model Downloads |
| |
|
| | You can download Ling-1T from the following table. If you are located in mainland China, we also provide the model on ModelScope.cn to speed up the download process. |
| |
|
| | <center> |
| |
|
| | | **Model** | **Context Length** | **Download** | |
| | | :-------: | :----------------: | :-------------------------------------------------------------------------------------------------------------------------------------------: | |
| | | Ling-1T | 32K -> 128K (YaRN) | [🤗 HuggingFace](https://huggingface.co/inclusionAI/Ling-1T) [🤖 ModelScope](https://www.modelscope.cn/models/inclusionAI/Ling-1T) | |
| |
|
| | </center> |
| |
|
| | Note: If you are interested in previous version, please visit the past model collections in [Huggingface](https://huggingface.co/inclusionAI) or [ModelScope](https://modelscope.cn/organization/inclusionAI). |
| |
|
| |
|
| | ## Quickstart |
| |
|
| | ### 🚀 Try Online |
| |
|
| | You can experience Ling-1T online at: [ZenMux](https://zenmux.ai/inclusionai/ling-1t?utm_source=hf_inclusionAI) |
| |
|
| | ### 🔌 API Usage |
| |
|
| | You can also use Ling-1T through API calls: |
| |
|
| | ```python |
| | from openai import OpenAI |
| | |
| | # 1. Initialize the OpenAI client |
| | client = OpenAI( |
| | # 2. Point the base URL to the ZenMux endpoint |
| | base_url="https://zenmux.ai/api/v1", |
| | # 3. Replace with the API Key from your ZenMux user console |
| | api_key="<your ZENMUX_API_KEY>", |
| | ) |
| | |
| | # 4. Make a request |
| | completion = client.chat.completions.create( |
| | # 5. Specify the model to use in the format "provider/model-name" |
| | model="inclusionai/ling-1t", |
| | messages=[ |
| | { |
| | "role": "user", |
| | "content": "What is the meaning of life?" |
| | } |
| | ] |
| | ) |
| | |
| | print(completion.choices[0].message.content) |
| | ``` |
| |
|
| | ### 🤗 Hugging Face Transformers |
| |
|
| | Here is a code snippet to show you how to use the chat model with `transformers`: |
| |
|
| | ```python |
| | from transformers import AutoModelForCausalLM, AutoTokenizer |
| | |
| | model_name = "inclusionAI/Ling-1T" |
| | |
| | model = AutoModelForCausalLM.from_pretrained( |
| | model_name, |
| | dtype="auto", |
| | device_map="auto", |
| | trust_remote_code=True, |
| | ) |
| | tokenizer = AutoTokenizer.from_pretrained(model_name) |
| | |
| | prompt = "Give me a short introduction to large language models." |
| | messages = [ |
| | {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"}, |
| | {"role": "user", "content": prompt} |
| | ] |
| | text = tokenizer.apply_chat_template( |
| | messages, |
| | tokenize=False, |
| | add_generation_prompt=True |
| | ) |
| | model_inputs = tokenizer([text], return_tensors="pt", return_token_type_ids=False).to(model.device) |
| | |
| | generated_ids = model.generate( |
| | **model_inputs, |
| | max_new_tokens=512 |
| | ) |
| | generated_ids = [ |
| | output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids) |
| | ] |
| | |
| | response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] |
| | ``` |
| |
|
| | ### 🤖 ModelScope |
| |
|
| | If you're in mainland China, we strongly recommend you to use our model from 🤖 <a href="https://modelscope.cn/models/inclusionAI/Ling-1T">ModelScope</a>. |
| |
|
| | ## Deployment |
| |
|
| | ### vLLM |
| |
|
| | vLLM supports offline batched inference or launching an OpenAI-Compatible API Service for online inference. |
| |
|
| | #### Environment Preparation |
| |
|
| | ```bash |
| | pip install vllm==0.11.0 |
| | ``` |
| |
|
| | #### Offline Inference: |
| |
|
| | ```python |
| | from transformers import AutoTokenizer |
| | from vllm import LLM, SamplingParams |
| | |
| | tokenizer = AutoTokenizer.from_pretrained("inclusionAI/Ling-1T") |
| | |
| | sampling_params = SamplingParams(temperature=0.7, top_p=0.8, repetition_penalty=1.05, max_tokens=16384) |
| | |
| | llm = LLM(model="inclusionAI/Ling-1T", dtype='bfloat16', trust_remote_code=True) |
| | prompt = "Give me a short introduction to large language models." |
| | messages = [ |
| | {"role": "system", "content": "You are Ling, an assistant created by inclusionAI"}, |
| | {"role": "user", "content": prompt} |
| | ] |
| | |
| | text = tokenizer.apply_chat_template( |
| | messages, |
| | tokenize=False, |
| | add_generation_prompt=True |
| | ) |
| | outputs = llm.generate([text], sampling_params) |
| | |
| | ``` |
| |
|
| | #### Online Inference: |
| |
|
| | ```bash |
| | vllm serve inclusionAI/Ling-1T \ |
| | --tensor-parallel-size 32 \ |
| | --pipeline-parallel-size 1 \ |
| | --trust-remote-code \ |
| | --gpu-memory-utilization 0.90 |
| | |
| | # This is only an example, please adjust arguments according to your actual environment. |
| | ``` |
| |
|
| | To handle long context in vLLM using YaRN, we need to follow these two steps: |
| | 1. Add a `rope_scaling` field to the model's `config.json` file, for example: |
| | ```json |
| | { |
| | ..., |
| | "rope_scaling": { |
| | "factor": 4.0, |
| | "original_max_position_embeddings": 32768, |
| | "type": "yarn" |
| | } |
| | } |
| | ``` |
| | 2. Use an additional parameter `--max-model-len` to specify the desired maximum context length when starting the vLLM service. |
| |
|
| | For detailed guidance, please refer to the vLLM [`instructions`](https://docs.vllm.ai/en/latest/). |
| |
|
| |
|
| | ### SGLang |
| |
|
| | #### Environment Preparation |
| |
|
| | We will later submit our model to SGLang official release, now we can prepare the environment following steps: |
| | ```shell |
| | pip3 install sglang==0.5.2rc0 sgl-kernel==0.3.7.post1 |
| | ``` |
| | You can use docker image as well: |
| | ```shell |
| | docker pull lmsysorg/sglang:v0.5.2rc0-cu126 |
| | ``` |
| | Then you should apply patch to sglang installation: |
| | ```bash |
| | # patch command is needed, run `yum install -y patch` if needed |
| | patch -d `python -c 'import sglang;import os; print(os.path.dirname(sglang.__file__))'` -p3 < inference/sglang/bailing_moe_v2.patch |
| | ``` |
| |
|
| | #### Run Inference |
| |
|
| | BF16 and FP8 models are supported by SGLang now, it depends on the dtype of the model in ${MODEL_PATH}. They both share the same command in the following: |
| | |
| | - Start server: |
| | ```bash |
| | python -m sglang.launch_server \ |
| | --model-path $MODEL_PATH \ |
| | --host 0.0.0.0 --port $PORT \ |
| | --trust-remote-code \ |
| | --attention-backend fa3 |
| | |
| | # This is only an example, please adjust arguments according to your actual environment. |
| | ``` |
| | MTP is supported for base model, and not yet for chat model. You can add parameter `--speculative-algorithm NEXTN` |
| | to start command. |
| | |
| | - Client: |
| | |
| | ```shell |
| | curl -s http://localhost:${PORT}/v1/chat/completions \ |
| | -H "Content-Type: application/json" \ |
| | -d '{"model": "auto", "messages": [{"role": "user", "content": "What is the capital of France?"}]}' |
| | ``` |
| | |
| | More usage can be found [here](https://docs.sglang.ai/basic_usage/send_request.html) |
| | |
| | |
| | |
| | ## Limitations & Future Plans |
| | |
| | While **Ling-1T** has made strong progress in efficient reasoning, cross-domain generalization, and training efficiency, several limitations remain: |
| | |
| | * **GQA-based attention**: stable for long-context reasoning but relatively costly. Future versions will adopt **hybrid attention** to improve efficiency. |
| | * **Limited agentic ability**: current model has room to grow in multi-turn interaction, long-term memory, and tool use. |
| | * **Instruction and identity issues**: occasional deviations or role confusion may occur; future updates will enhance **alignment and consistency**. |
| | |
| | The future versions of Ling-1T will continue to evolve in architecture, reasoning, and alignment, advancing the series toward more general intelligence. |
| | |
| | |
| | ## License |
| | |
| | This code repository is licensed under [the MIT License](https://github.com/inclusionAI/Ling-V2/blob/main/LICENSE). |
