Update model card: fix SGLang heading, add Model Information, license, hardware, and inference settings

README.md

@@ -3,6 +3,8 @@ language:
 - en
 - zh
 license: other
+license_name: agnes-ai-terms-of-use
+license_link: https://agnes-ai.com/
 pipeline_tag: image-text-to-text
 tags:
 - agnes-ai
@@ -76,13 +78,49 @@ Agnes 2.5 Pro Alpha is evaluated against the same comparison set selected for Or
 † GDPval-AA v2 uses Artificial Analysis' normalized score, <code>(Elo − 500) / 2000</code>. Non-Hallucination Rate is <code>1 − hallucination rate</code>. Higher is better for every benchmark. Snapshot checked August 18, 2026; values may change as evaluations are updated.
 </p>
 
+## Model Information
+
+| Property | Value |
+|---|---|
+| Developed by | Agnes AI |
+| Model name | Agnes 2.5 Pro Alpha |
+| Model type | Proprietary multimodal reasoning model |
+| Languages | English, Chinese |
+| Context window | 1,048,576 tokens |
+| Maximum output | 65,536 tokens |
+| Input modalities | Text, image |
+| Output modality | Text |
+| Precision | BF16 |
+| Tool calling | Yes |
+| Streaming | Yes |
+| Release date | July 2026 |
+
+## License
+
+This model is released under the **Agnes AI Terms of Use** (`license: other`). It is not licensed as Apache-2.0, MIT, or any other OSI open-source license.
+
+Use of the weights in this repository and of the hosted API is subject to the [Agnes AI](https://agnes-ai.com/) terms. For commercial licensing questions, contact [support@agnes-ai.com](mailto:support@agnes-ai.com).
+
+## Hardware Requirements
+
+The Quickstart launch command uses 8-GPU tensor parallelism. The checkpoint is a large multi-shard BF16 package; a single GPU is not sufficient.
+
+| Resource | Recommendation |
+|---|---|
+| GPUs | 8× NVIDIA H200 (141 GB) or equivalent |
+| Tensor parallel | `--tp 8` |
+| Host memory / disk | Fast NVMe with about **1 TB** free for weights, tokenizer files, and download cache |
+| Context length | The sample command sets `--context-length 1024000`. If you hit out-of-memory errors, lower this value |
+| Network | Optional. The same model is also served at `https://apihub.agnes-ai.com/v1` without local GPUs |
+
 ## Quickstart
 
 <div style="border-left:4px solid #3248AF;background:rgba(50,72,175,.08);border-radius:6px;padding:12px 16px;font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;font-size:14px;line-height:1.6">
 <div style="font-weight:700;color:#3248AF;margin-bottom:6px">REASONING MODEL</div>
 <p style="margin:0"><b>Agnes 2.5 Pro Alpha</b> uses extended reasoning for complex tasks and is available through OpenAI-compatible Chat Completions and Responses APIs. Keep API keys in environment variables and use publicly accessible URLs for image inputs.</p>
 </div>
-### SGLang
+
+<h3>SGLang</h3>
 
 ```bash
 python -m sglang.launch_server \
@@ -180,6 +218,20 @@ curl https://apihub.agnes-ai.com/v1/responses \
   }'
 ```
 
+## Recommended Inference Settings
+
+Use sampling rather than greedy decoding. Leave enough `max_tokens` / `max_output_tokens` for extended reasoning.
+
+| Setting | Default | Coding and tool use |
+|---|---|---|
+| `temperature` | 1.0 | 0.2–0.3 |
+| `top_p` | 0.95 | 0.95 |
+| `top_k` | 20 | 20 |
+| `repetition_penalty` | 1.05 | 1.05 |
+| `max_tokens` | 2000 or higher | 2000 or higher |
+
+The coding examples in Quickstart use `temperature` 0.2 (Chat Completions) and 0.3 (Python). Raise `max_tokens` if a response stops early.
+
 ## Model Capabilities
 
 | Capability | Support |