model.output_truncated
The answer was cut at the method's token cap
| Type | https://mthds.ai/errors/model.output_truncated |
| Family | model: The model a step names. |
| Domain | method: the method's author acts, by changing the method |
| Retry | no: repeating the operation fails the same way |
| HTTP status | 422 |
| Raised as | a problem document, sent as the body of an HTTP response or kept in a run's record |
What it means
The model stopped at the step's max_tokens before finishing its answer.
What to do
Raise the step's max_tokens, or change its prompt so that it asks for a shorter answer.
Next steps
A problem carries the step aimed at whoever reads it: the first when its reader is the domain's actor, the method's author, and the second for any other reader.
| Reader | Kind | Detail |
|---|---|---|
| The domain's actor | change_method |
Raise the step's max_tokens above {max_tokens}, or ask for a shorter answer, then run the method again. |
| Any other reader | contact_owner |
Ask the method's owner to raise the step's token cap. |
Detail
The problem's detail is written from this template, filled with the occurrence's parameters:
The model '
{model}' stopped at the step's cap of{max_tokens}tokens, after{output_tokens}tokens.
Parameters
Every parameter is present on every occurrence of the code.
| Name | Type | Description |
|---|---|---|
model |
identifier |
The model. |
max_tokens |
integer |
The cap the step sets. |
output_tokens |
integer |
The tokens the model produced. |
See also
- Model References
- Error Registry, which describes the members of this page
- Error Codes, every code of the core registry