Update README with latest models (#218)

Co-authored-by: cheahjs <818368+cheahjs@users.noreply.github.com>
This commit is contained in:
github-actions[bot]andcheahjs authored and GitHub committed 2025-08-08 09:21:33 +01:00
1 parent d6ed99e025
commit ccb308c9b7
1 file changed
+5
+5
View File
@@ -101,6 +101,7 @@ Models share a common quota.
- [qwen/qwen3-30b-a3b:free](https://openrouter.ai/qwen/qwen3-30b-a3b:free)
- [qwen/qwen3-4b:free](https://openrouter.ai/qwen/qwen3-4b:free)
- [qwen/qwen3-8b:free](https://openrouter.ai/qwen/qwen3-8b:free)
- [qwen/qwen3-coder:free](https://openrouter.ai/qwen/qwen3-coder:free)
- [sarvamai/sarvam-m:free](https://openrouter.ai/sarvamai/sarvam-m:free)
- [tencent/hunyuan-a13b-instruct:free](https://openrouter.ai/tencent/hunyuan-a13b-instruct:free)
- [thudm/glm-z1-32b:free](https://openrouter.ai/thudm/glm-z1-32b:free)
@@ -276,6 +277,10 @@ Extremely restrictive input/output token limits.
- OpenAI GPT-4o mini
- OpenAI Text Embedding 3 (large)
- OpenAI Text Embedding 3 (small)
- OpenAI gpt-5
- OpenAI gpt-5-chat (preview)
- OpenAI gpt-5-mini
- OpenAI gpt-5-nano
- OpenAI o1
- OpenAI o1-mini
- OpenAI o1-preview