-
Notifications
You must be signed in to change notification settings - Fork 1
Permalink
Choose a base ref
{{ refName }}
default
Choose a head ref
{{ refName }}
default
Comparing changes
Choose two branches to see what’s changed or to start a new pull request.
If you need to, you can also or
learn more about diff comparisons.
Open a pull request
Create a new pull request by comparing changes across two branches. If you need to, you can also .
Learn more about diff comparisons here.
base repository: thisisartium/gpt_agent
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v9.1.0
Could not load branches
Nothing to show
Loading
Could not load tags
Nothing to show
{{ refName }}
default
Loading
...
head repository: thisisartium/gpt_agent
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: main
Could not load branches
Nothing to show
Loading
Could not load tags
Nothing to show
{{ refName }}
default
Loading
- 9 commits
- 14 files changed
- 1 contributor
Commits on Feb 22, 2024
-
Retry runs on time/token based rate limiting
When the run status indicates that OpenAI has rate-limited the request due to tokens per minute or requests per minute, we will retry the requests after a configurable delay and up to a configurable number of times (defaults to no retries). With this commit, we slightly modify the error code returned with the RunFailed event to indicate whether the failure is due to temporary rate limiting or quota being reached as well as whether a retry will be attempted. We return the event even when a retry will be attempted, so that the subscribers can be aware that rate-limiting is occuring.
Configuration menu - View commit details
-
Copy full SHA for df72f29 - Browse repository at this point
Copy the full SHA df72f29View commit details -
Configuration menu - View commit details
-
Copy full SHA for c40adea - Browse repository at this point
Copy the full SHA c40adeaView commit details -
Add specific events for rate limiting errors
Instead of forcing client code to evaluate the content of the code field on a RunFailed event, we introduce three new event types that can be used for pattern matching in subscribers: * RateLimited - means that OpenAI has rate limited the organization based on either the number of requests or the number of tokens per minute. Includes a `retries_remaining` field. * RateLimitRetriesExhausted - means that we were rate-limited as per above and that the run will not be automatically retried. * OrganizationQuotaExceeded - means that the organizations monthly spending limit on tokens has been reached or exceeded, and the system cannot process any more runs until either the limit is lifted or the billing period resets.
Configuration menu - View commit details
-
Copy full SHA for 6390ce3 - Browse repository at this point
Copy the full SHA 6390ce3View commit details
Commits on Feb 23, 2024
-
Fix bug due to not setting running? false
I missed that I needed to set the state of the agent to "not running" after handling failed runs that will not be retried but there were two cases where I was not.
Configuration menu - View commit details
-
Copy full SHA for 5ae9a41 - Browse repository at this point
Copy the full SHA 5ae9a41View commit details
Commits on Mar 1, 2024
-
Prevent hang due to OpenAI receive time
We now no longer allow the receive timeout to effectively be longer than the agent timeout period. This ensures that we won't hang "forever" waiting on OpenAI to respond.
Configuration menu - View commit details
-
Copy full SHA for c8bd228 - Browse repository at this point
Copy the full SHA c8bd228View commit details
Commits on Mar 2, 2024
-
Fix bug handling error response in tool submission
If openai returns an error response when we submit tool output, we will retry up to 2 more times and then publish a failure event.
Configuration menu - View commit details
-
Copy full SHA for d1c08c0 - Browse repository at this point
Copy the full SHA d1c08c0View commit details
Commits on Oct 8, 2024
-
Configuration menu - View commit details
-
Copy full SHA for e9829d7 - Browse repository at this point
Copy the full SHA e9829d7View commit details -
Configuration menu - View commit details
-
Copy full SHA for c168b4c - Browse repository at this point
Copy the full SHA c168b4cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 84e4a46 - Browse repository at this point
Copy the full SHA 84e4a46View commit details
Loading
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff v9.1.0...main