Update app.py

Update app.py

feat(models): add Granite-4.0-Micro and Qwen3-4B-Instruct-2507 to MODELS registry

feat(models): Added three new models

add 5 models from liquid ai

add smollm2 135m multilingual

add parser_model_ner_gemma_v0 based on gemma 3 370m it

Add Gemma-3-270m Taiwan

add gemma-3-270m-it

remove prevously added breeze models (as it didn't work), add smollm 135m taiwan

add breeze models

add 3 sub-1B TW models from ShengweiPeng

add Qwen2.5-Taiwan-3B-Reason-GRPO & Llama-3.2-Taiwan-1B

disable cache for better compability

make qwen2.5-taiwan 1.5b the default

upgrade transforemrs to 4.53.3+

upgrade transformers to 4.46.1+

add timm to requirements

add access_token to model loader

add gemma 3n, smallthinker 4b0.6b and falcon h1-1.5b

rename duckduckgo_search to ddgs

remove qwen3 30ba3b & 32b, but add 14b

use fp8 version of qwen3 30ba3b & 32b

add autoawq

add qwe3 30ba3b awq & 30b awq

update prompt

update prompt

fix err in prompt

improve prompt for better web search result merging

Add D-R1-0528 8B, Nemotron 1.5B, Qwen2.5 Taiwan 1.5B

reduce rep penality to 1.2

increase rep penality as on many slms, reptition is frequent.

revert device_map to audo

Add smolLM2-135M GRPO and SmolLM2-360M

Update README.md

update readme

add Qwen2.5-Omni-3B & MiMo-7B-RL

disable models unrunnable on hf spaces

switch Phi-4-mini-Instruct" from unsloth to microsoft

extend max tokens

adjust layout

add "Granite-4.0-Tiny-Preview" model

add phi-4 reasonning and phi-4-mini reasonning

bugfix: set device to xpu by mistake

add all qwen3 variants

user can define search timeout

give 5 second for web earch to gather reults

support thinking models and streamingly display thought

do not preview prompt at error return from chat response

inject assistant placeholder at right time