Tomáš Dvořák comments

Results 140 comments of


                                            Tomáš Dvořák

Empty output from chat-endpoint / non-empty endpoint for non-chat endpoint

Great work! I can acknowledge that the issue has been fixed in v0.0.53

Cannot parse header and one column contain DOM elements.

Can I see such HTML Table so I can debug it?

Cannot parse header and one column contain DOM elements.

If your table looks like this ``` A C A1 B1 C1 A1 C1 ``` Then you can do this ```typescript const data = await tableParser(page, { selector: '#table-overview', asArray:...

fix: example -updates the examples to the new llama 3.2 3B instruct

I am closing this for being stale.

fix: 'that' is not defined

ping

[feat] use idiomatic spans in beeai instrumentation

We reworked the instrumentation package for BeeAI. Is this issue still present?

[feat] use idiomatic spans in beeai instrumentation

The `start/finish` events should not be present in the latest version `0.1.10`.

[feat] use idiomatic spans in beeai instrumentation

Can you clarify a little bit, @codefromthecrypt?

[feat] use idiomatic spans in beeai instrumentation

We support multiple providers, and we internally rely on LiteLLM, and we can't drop it. Only for the OpenAI adapter can we do that, but I'm not sure if it...

Update default examples to llama3.2

Great @Abiji-2020! Firstly you need to define the model template in `src/adapters/shared/llmChatTemplates.ts`. then you have to updates presets for BAM and WatsonX, finally update the examples.