Olivier Chafik
Olivier Chafik
Once you open the dialog and clear the list of materials / leave it empty + export / save the setting, it shouldn’t prompt for it when exporting to 3mf....
@pks5 How are you building your iframe URLs? The list of materials (undefined at first, then empty array if the dialog is closed/saved without adding a color) is encoded in...
@yhirose What do you think about this approach?
> What's the reason for making a different release for each arch? @slaren Building for a single arch seems a lot faster, and having separate artefacts instead of (cuda-)fat binaries...
I'm slightly worried about allowing message content array w/o requiring a strict message role alternance. And very worried about the breaking change. Most inference APIs (OpenAI's chat completions, Claude's, but...
FYI I've updated (my fork of) Minja w/ [support for GLM 4.6's template](https://github.com/ochafik/minja/pull/5). Might affect how you deal w/ the polyfills, as it should now detect GLM's tool call capability...
> Currently, the official Minimax-M2 chat template fails to run tool calls because dict.items() and list[-1] are not supported by llama.cpp’s Jinja2 rendering engine. @hksdpc255 Both should be supported. The...
Hey @henryclw, thanks for filing this! Note that some models generate a tool call id on their own (e.g. Cohere Command R7B, Mistral Nemo). Could you check if langchain requires...
@Sherlock-Holo Thanks for filing this! FYI, this is currently halfway between working as intended (by the DeepSeek R1 template authors - both the *latest* official DS template and our `llama-cpp-deepseek-r1.jinja`...
> [@ochafik](https://github.com/ochafik) did you mean use the official template or `llama-cpp-deepseek-r1.jinja` will cause the first `` tag miss? @Sherlock-Holo Both, actually! Luckily the GGUF from bartowski was converted before DeepSeek...