[Bug]: Can not run JS sample "03.ai-concepts/e.customModel-LLAMA"

[frankqianms]

Language

Javascript/Typescript

Version

JS 1.3.0

Description

Run sample https://github.com/microsoft/teams-ai/tree/main/js/samples/03.ai-concepts/e.customModel-LLAMA and got the following error in log.

CHAT PROMPT: [ { "role": "system", "content": "The following is a conversation with an AI assistant.\nThe assistant can turn a light on or off. The assistant can only manage lights. The assistant can only discuss lights.\n\ncontext:\nThe lights are currently off.\n\nactions:\n LightsOn:\n description: Turns on the lights\n LightsOff:\n description: Turns off the lights\n\n\nUse the actions above to create a plan in the following JSON format:\n{\"type\":\"plan\",\"commands\":[{\"type\":\"DO\",\"action\":\"\",\"parameters\":{\"\":}},{\"type\":\"SAY\",\"response\":\"\"}]}" }, { "role": "user", "content": "hello" }, { "role": "assistant", "content": "" }, { "role": "user", "content": "Return a JSON object that uses the SAY command to say what you're thinking." }, { "role": "assistant", "content": "" }, { "role": "user", "content": "Return a JSON object that uses the SAY command to say what you're thinking." }, { "role": "assistant", "content": "" }, { "role": "user", "content": "Return a JSON object that uses the SAY command to say what you're thinking." } ] CHAT RESPONSE: status: 200 undefined

[onTurnError] unhandled error: Error: Reached max model response repair attempts. Last feedback given to model: "Return a JSON object that uses the SAY command to say what you're thinking." Error: Reached max model response repair attempts. Last feedback given to model: "Return a JSON object that uses the SAY command to say what you're thinking." at LLMClient.repairResponse (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\LLMClient.ts:488:24) at processTicksAndRejections (node:internal/process/task_queues:95:5) at async LLMClient.repairResponse (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\LLMClient.ts:495:16) at async LLMClient.repairResponse (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\LLMClient.ts:495:16) at async LLMClient.completePrompt (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\LLMClient.ts:347:28) at async ActionPlanner.completePrompt (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\ActionPlanner.ts:250:16)
at async ActionPlanner.continueTask (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\ActionPlanner.ts:178:24) at async ActionPlanner.beginTask (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\planners\ActionPlanner.ts:153:16) at async AI.run (C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\AI.ts:367:28) at async C:\Users\frankqian\TeamsApps\e.customModel-LLAMA\node_modules\@microsoft\teams-ai\src\Application.ts:823:21

Reproduction Steps

1.Follow all steps mentioned in ReadMe
2.Got the error
3.
...

[corinagum]

This is a duplicate of: https://github.com/microsoft/teams-ai/issues/1522

Ideally the sample needs to be updated to a non-sequence Augmentation and non-monologue augmentation sample. For example, if you were to update the code to match our twenty questions sample, the sample would work better. we just don't have bandwidth at the moment to update the sample.

If you do end up making these changes, please also file a PR!

[frankqianms]

This is a duplicate of: #1522

Ideally the sample needs to be updated to a non-sequence Augmentation and non-monologue augmentation sample. For example, if you were to update the code to match our twenty questions sample, the sample would work better. we just don't have bandwidth at the moment to update the sample.

If you do end up making these changes, please also file a PR!

Got it! Thanks for sharing the information.