[✅ Solved] The message always breaks off and doesn't have it's ending

[drowsy laurel]

Just like that, for example(watch screenshot)

[timid cipher]

It looks like you're experiencing an issue where the message cuts off before reaching its end. This is typically caused by the model reaching its maximum token limit per response. To resolve this, you can try one of the following:

1. Increase Context Limit: Use a model with a larger context limit if available. On Pawan.Krd, for more tokens per message, consider using models like CosmosRP-Pro-V3 which supports up to 32k tokens for supporters.

2. Reduce Message Length: Limit the length of each message or dialogue in your prompts to avoid reaching the token limit quickly.

3. Segment Longer Texts: Break down longer messages into smaller segments and send them as multiple prompts to ensure they complete without being cut off.

If the issue persists despite trying these suggestions, and especially if you're using a model with a high token limit, consider checking the usage rate and settings to ensure everything is configured correctly. If necessary, create a support ticket for further investigation through our Create Ticket Channel.

Please note: This is an automated response from @timid cipher. While it strive to provide accurate assistance, It may occasionally make mistakes. If you find any inaccuracies or need further clarification, feel free to wait for our community helpers to give you further guidance. Thank you.

[drowsy laurel]

It doesn't even matter if i reduce the "max new token", it just keeps breaking off

My settings

[hearty sparrow]

Reducing max tokens make the response shorter and not finish, you want it higher, preferably the max you can make it.

Chub has many issues regarding the response being cut off in general

[drowsy laurel]

Well, imma just try janitor instead then

[timid cipher]

[✅ Solved] The message always breaks off and doesn't have it's ending