v2.5.0

Laxcorp-Research/project-ravenv2.5.0Sep 23, 2026by claxman

Upgrading?

AI Summary

This release introduces the ability to resume previous recording sessions to maintain context across multiple days, while also fixing critical issues preventing OpenAI models from working with GPT-5.x.

Key Highlights

  • Resume a session: Continue recording into a previous session to keep transcript, notes, and context intact.
  • OpenAI compatibility: Fixed API requests to use `max_completion_tokens` instead of `max_tokens` for GPT-5.x.
  • Key validation: Fixed rate limiting issues when saving both OpenAI and Anthropic keys simultaneously.
  • Effort picker: Removed 'Max' option for GPT-5.6 and added a fallback to 'Low' for existing users.

New Features

  • Resume a session: Open a saved session in the dashboard and click Resume to keep recording into it, carrying over the transcript, answers, notes, and chat history.
  • Session Memory Storage: Assist's compacted session memory is now stored on the session row (local SQLite) to enable seamless resumption.

Full Release Notes

## New

- **Resume a session.** Open a saved session in the dashboard and click **Resume** to keep recording into it. The transcript, answers, notes, Ask chat and session id all carry over, and Assist is handed what it knew from the earlier sitting, so a two-day interview stays one session and the model answers day two with day one in context. Duration counts recorded time only, the session shows how many sittings it has, and the transcript marks where each resume happened. Assist's compacted session memory is now stored on the session row (local SQLite, deleted with the session) to make this work.

## Fixes

- **OpenAI models work again.** Every OpenAI request failed with `400 Unsupported parameter: 'max_tokens'`. Requests now send `max_completion_tokens`, which GPT-5.x requires. Fixes #30.
- **OpenAI key can be saved alongside an Anthropic key.** Settings validated the two keys in separate calls and the second was rate-limited, leaving "Please wait a moment and try again" and a spinner that never stopped. Both keys now validate in one call.
- **Effort picker no longer offers Max on GPT-5.6.** The API rejects it; anyone who had selected it was getting a 400 on every request. Existing Max selections fall back to Low.

## Install

- **macOS 12+ (Apple Silicon):** [Raven-Mac-2.5.0-Installer.dmg](https://github.com/Laxcorp-Research/project-raven/releases/download/v2.5.0/Raven-Mac-2.5.0-Installer.dmg)
- **Windows 10/11 (x64):** [Raven-Windows-2.5.0-Setup.exe](https://github.com/Laxcorp-Research/project-raven/releases/download/v2.5.0/Raven-Windows-2.5.0-Setup.exe) (attached once the Windows build finishes)

Existing installs on 2.4.1 or later update in place: **Update now** downloads, **Restart & update** installs. Your keys and history are untouched.

Related Repositories