Multi-turn UX: new-chat button; example click resets conversation; 1k-token sliding-window history; 4th example (sleep) a284b48 verified codelion commited on May 26
INT8 dynamic quant (~2x CPU, verified all 3 modes); add self-spec to chat mode selector 873a061 verified codelion commited on May 26
Compare/chat prompts: drop count-based "three tips" (looks truncated at 48 tok); use self-contained short-answer prompts (ocean / neural net / exercise) 0249a53 verified codelion commited on May 26
Fix chat crash: gradio 6 Chatbot content is a list of parts; flatten to text before apply_chat_template 1ed429f verified codelion commited on May 26
Fix gradio 6.14: gr.Chatbot dropped the type= arg (messages is default) f411959 verified codelion commited on May 26
Custom chat layout: compact mode/max-tokens/temp controls above input; single-column examples (readable); header link to codelion/dhara-250m df2112c verified codelion commited on May 26
Disable example caching (HF Spaces defaults cache_examples=True) so example clicks generate live, not pre-baked 4681623 verified codelion commited on May 26
Coherence: rep-penalty 1.3 in diffusion/self-spec (no more collapse); clip to last full sentence (no mid-word cutoff); exposed config panel (mode/max-tokens/temperature); greedy AR in compare; better default prompts 7c0285b verified codelion commited on May 26
Fix gradio 6 build: examples as list-of-lists (additional_inputs); move theme/css to launch() cf976b5 verified codelion commited on May 26
Chat mode toggle (AR/diffusion w/ live ▓▒░ denoise); streaming 3-mode compare reveal; max_new 128 + no_repeat_ngram 732ea5f verified codelion commited on May 26
Fixed-height scrollable chat; pass attention_mask (silence pad==eos warning); _enc helper a36cfa9 verified codelion commited on May 26
Denoiser-terminal UI (mask-glyph aesthetic, amber CRT) + streaming per-mode compare with progress 32de77f verified codelion commited on May 26
Link model (models: codelion/dhara-250m) for Spaces-using-this-model widget 3ecfd57 verified codelion commited on May 26
Add torch to requirements (gradio SDK base has no torch) 7638ff6 verified codelion commited on May 26
gradio SDK 6.14.0 (proven to import with hub 1.16) + transformers 5.5 + messages-format app 51b846a verified codelion commited on May 26
Pin gradio==4.44.1 + use 4.x tuples ChatInterface API (model already loads) 1a232f4 verified codelion commited on May 26
Switch to Docker SDK; pip-resolved transformers==5.5.0 + gradio (fixes tokenizer/HfFolder) a54a16a verified codelion commited on May 26
Pin coherent stack (py3.12, gradio4.44, transformers<5, hub<1) to fix HfFolder/audioop; title ChatDhara a113c18 verified codelion commited on May 26
Fix py3.13 gradio import: add audioop-lts (audioop removed in 3.13) e2c8b3f verified codelion commited on May 26