Update README.md

README.md

@@ -7,7 +7,9 @@ license: mit
 <p align="center">🤗 <a href="https://huggingface.co/inclusionAI">Hugging Face</a>&nbsp;&nbsp; | &nbsp;&nbsp;🤖 <a href="https://modelscope.cn/organization/inclusionAI">ModelScope </a>&nbsp;&nbsp; | &nbsp;&nbsp;🐙 <a href="https://openrouter.ai/inclusionai/ling-3.0-tiny:free">OpenRouter </a>&nbsp;&nbsp;</p>
 
 # Introduction
-We are introducing Ling-3.0-tiny, a native hybrid reasoning model with 7.9B total parameters and only 1.3B activated parameters during inference. This release provides BF16, FP8, and INT4 variants for different hardware, performance, and deployment requirements.
+We are introducing **Ling-3.0-tiny**, a lightweight hybrid reasoning MoE model with **7.9B** total parameters and only **1.3B** activated parameters per token. 
+It is designed to deliver strong reasoning and agentic capabilities under a small inference compute footprint, making advanced model capabilities more accessible for local and resource-constrained deployment. 
+BF16, FP8, and INT4 weights are provided for a wide range of hardware and deployment settings.
 
 Key highlights of the model are summarized below:
 
@@ -16,13 +18,13 @@ Key highlights of the model are summarized below:
 + **Local and Edge Deployment:** Designed for efficient local deployment, Ling-3.0-tiny has been validated on **NVIDIA DGX Spark, Apple Silicon MacBook, and Mac mini**, enabling capable reasoning and agentic workloads without datacenter-class GPUs. With FP8, Ling-3.0-tiny reaches around **100-105 tokens/s on DGX Spark** and **86-90 tokens/s on an M4 Pro MacBook**, with approximately **8.34 GiB peak memory usage** at an 8K context length.
 
 # Model Overview
-Ling-3.0-tiny adopts the native hybrid linear attention architecture of Ling-3.0 series and is further optimized for lightweight, accessible deployment scenarios. The model has 7.9B total parameters, with only 1.3B parameters activated per token.
+Ling-3.0-tiny inherits the hybrid linear attation architecture of Ling-3.0 series, while being specifically optimized for lightweight and accessible deployment. The model has 7.9B total parameters, with only 1.3B parameters activated per token.
 
 The architecture of Ling-3.0-tiny is not designed to optimize a single technical metric; instead, it aims to translate efficiency into real-world task performance.
 
 + A 3:1 KDA–MLA architecture (3 KDA layers and 1 MLA layer in every 4-layer block) enables more efficient long-context processing;
-+ A sparse MoE with 128 experts activates 8 routed experts and 1 shared expert per token, enabling the model to deliver more comprehensive capabilities with only 1.3B activated parameters;
-+ Native hybrid reasoning enables fast responses for routine tasks and multi-step reasoning for complex tasks within one single model.
++ A sparse MoE with 128 experts activates 8 routed experts and 1 shared expert per token, enabling broad model capabilities with only 1.3B activated parameters per token.
++ Native hybrid reasoning enables fast responses for routine tasks and multi-step reasoning for complex tasks within a single model.
 
 Together, these designs target three goals:
 
@@ -34,15 +36,17 @@ Together, these designs target three goals:
 ![](https://intranetproxy.alipay.com/skylark/lark/0/2026/png/202156655/1786334509742-e888d58f-c9af-44d0-a64a-a9b6ba0cb652.png)
 
 # Evaluation
-We evaluated Ling-3.0-tiny across general agentic tasks, coding tasks, long-context understanding, knowledge reliability, mathematical and scientific reasoning, and instruction following.
-
-Ling-3.0-tiny scores **25** on the Artificial Analysis Intelligence Index v4.1.1 and **16** on the Artificial Analysis Agentic Index. In Artificial Analysis testing, the model achieves an output speed of more than **160 tokens/s** and an end-to-end latency of approximately **18 seconds** for generating 500 tokens (including the reasoning model's thinking time), reflecting a balance among overall capability, activated parameter scale, and response efficiency.
+We evaluated Ling-3.0-tiny across agentic tasks, coding, long-context understanding, knowledge reliability, mathematical and scientific reasoning, and instruction following.
+Ling-3.0-tiny achieves a score of **25** on the Artificial Analysis Intelligence Index v4.1.1 and **16** on the Artificial Analysis Agentic Index. 
+In Artificial Analysis testing, Ling-3.0-tiny reaches an output speed of over **160 tokens/s**, with approximately **18 seconds** of end-to-end latency for a 500-token response, including reasoning time. 
+These results highlight the model's efficiency relative to its 1.3B activated parameter footprint.
 
 The following table presents representative benchmarks for Ling-3.0-tiny:
 
 ![image](https://cdn-uploads.huggingface.co/production/uploads/6502cf8fbdaeae26417cd3c9/g9Thw4ohjkDGYw0Cq7mKi.png)
 
-> All Ling-3.0-tiny evaluation results use Thinking mode; the recommended sampling parameters are `temperature=1.0`, `top_p=0.95`, and `top_k=20`.
+> + All Ling-3.0-tiny evaluation results use Thinking mode; the recommended sampling parameters are `temperature=1.0`, `top_p=0.95`, and `top_k=20`.
+> + Terminal-Bench 2.1: Evaluated under the Artificial Analysis (AA) protocol using the default Terminus 2 harness, a unified 2-hour timeout, the provided JSON parser in preserve-thinking mode, and 3 runs per task (mean). Decoding uses temperature=1.0, max_new_tokens=32K, with a 256K context window.
 
 # Quickstart
 ## SGLang
@@ -98,7 +102,7 @@ curl -s http://localhost:30000/v1/chat/completions \
 For `--reasoning-parser ling3` / `--tool-call-parser ling3`, the HiCache + Mooncake L3 setup, and GSM8K / bench_serving reproduction commands, see the cookbook page linked above.
 
 ## vLLM
-### Install our vLLM
+### Install vLLM with Ling-3.0 Support
 ```bash
 pip install uv
 
@@ -114,7 +118,7 @@ VLLM_USE_PRECOMPILED=1 uv pip install --editable . --torch-backend=auto
 ```
 
 ### Run Inference
-Here is the example to run Ling-3.0-tiny with 1 GPU, where the server port is `${PORT}`:
+Here is the example to run Ling-3.0-tiny with a single GPU, where the server port is `${PORT}`:
 
 **Server**
 
@@ -134,7 +138,7 @@ vllm serve "$MODEL_PATH" \
 
 **Client**
 
-We recommend using the sampling parameters `temperature=1.0`, `top_p=0.95`, and `top_k=20`, and enabling `enable_thinking` for better performance.
+For better performance, We recommend setting `enable_thinking=true` with `temperature=1.0`, `top_p=0.95`, and `top_k=20`.
 
 ```bash
 curl -s http://${MASTER_IP}:${PORT}/v1/chat/completions \
@@ -150,8 +154,7 @@ curl -s http://${MASTER_IP}:${PORT}/v1/chat/completions \
 ```
 
 ## Ollama
-> This configuration has been verified on an M4 Pro Mac with 48 GB of unified memory.
->
+> + This configuration has been verified on an M4 Pro Mac with 48 GB of unified memory.
 
 ### Preparation and Build
 ```bash
@@ -164,10 +167,8 @@ cmake -B build .
 cmake --build build --parallel 8
 ```
 
-> Support is currently provided by [ollama/ollama#17643](https://github.com/ollama/ollama/pull/17643) and is limited to running via MLX on Apple Silicon.
->
-> Use the local `./ollama` executable built from source in this section. This functionality is not yet included in the official Ollama release.
->
+> + Support is currently provided by [ollama/ollama#17643](https://github.com/ollama/ollama/pull/17643) and is limited to running via MLX on Apple Silicon.
+> + Use the local `./ollama` executable built from source in this section. This functionality is not yet included in the official Ollama release.
 
 ### Import Model
 Replace `/absolute/path/to/bf16_weights` with the absolute path to the BF16 model weights directory. The imported model will be named `ling-tiny-bf16`: