19:47

• Triggered Tasks are here! Set up AI tasks that run automatically in the background when events fire.

21:10

• Triggered Tasks got a makeover: clearer Test and Add flows on the tasks page, target prompts are now required, and active triggers show up right in your prompt settings
• Fixed a bug where parent notes created from forwarded emails weren't correctly showing they had child notes attached

22:44

• Fixed bug Triggered Tasks weren't auto applying
• Fixed a bug where sub-note pills disappeared if the child's title contained an "@" symbol.

01:39

• You can now email notes directly to a specific prompt — each prompt gets its own email address in Settings
• Public pages now show the note title in the browser tab instead of a generic page title
• Backlink history toggle now works consistently across all views
• Grug tidied up the scrolls and sharpened the chisel (planning docs, changelog housekeeping)

02:21

• Fixed a bug where AI responses could occasionally cut out or error mid-stream on Speed model
• Grug tidied up the cave ledger (changelog housekeeping)

10:02

• New: graph view — explore how your notes connect to each other visually. Click around to see relationships between topics, people, and ideas. I don't even find this that useful lol, but it was easy to add.

02:38

• Default load state is no longer focused ie no active textarea     Press tab to focus and start typing.     Press space to activate/deactivate voice mode.
• Live audio stays in sync when your notes are manually edited
• Live voice editor now supports timers and has smarter guardrails to keep things on track
• Clicking the drag handle above each block moves the audio target (left border highlight) without activating the input/textarea
• Grug reorganized the cave notice board (changelog and error monitoring updates)

10:40

• Notes now load in read mode — no more keyboard popping up on page load. Press Tab to start typing, or Space to start dictating.
• You can press Space anywhere on the page to jump straight into live voice mode.
• Click on a note's handle area to set it as your voice dictation target.

14:17

• Live voice editing now stays in sync when your note is changed elsewhere (e.g. from another tab or device) — no more overwriting your latest edits
• GrugTimer now works during live voice sessions — set and manage timers hands-free while dictating
• Improved focus tracking in live audio mode so the editor doesn't lose your place
• Grug taught the cave parrot some manners so it won't go off-script mid-conversation (voice AI guardrails, error pipeline improvements)

14:26

• Fixed a bug where deleting an email agent didn't work properly
• Tasks page no longer jumps around on mobile, and pull-to-refresh won't interrupt you mid-scroll
• Landing page looks better on phones
• Your sidebar open/closed preference now sticks across sessions
• Audio editor is smarter about how it processes your recordings, ie some very slight prompt tuning
• Grug rearranged the bones in the cave and swept behind the fire pit (dependency cleanup, internal tooling, eval improvements)

17:34

• Fixed a sync issue where audio/edit-created notes wouldn't appear until the second refresh

00:42

Both commits are essentially internal: a refactor of AI prompt templates (deduplication into partials), improved agent chat preview cleaning, and a minor terminology rename ("heartbeat" → "check-in"). The changelog commit is just a changelog file. Here's the changelog:

• Agents now show "Last check-in" instead of "Last heartbeat" — clearer label for when your agent last ran
• Grug reorganized the scrolls in the prompt library and taught agents to summarize their past conversations more cleanly (internal refactoring, prompt deduplication)

21:09

• Lists now support stages — group items into columns like a board (e.g., To Do / In Progress / Done). Set them up in your prompt's attribute definitions.
• When you share a URL in chat, Grug can now fetch and read the page directly instead of doing a web search.
• Fixed a bug where note syncing could break or show stale content when editing across tabs.
• Agents with custom schedules now properly require you to pick at least one day. Unsubscribing from email delivery no longer disables agents that also deliver via Telegram or in-app.
• API and CLI access is now a paid feature.
• Grug reinforced the cave walls and patched a few leaky pipes (database connection cleanup, internal routing hardening, performance tuning).

Found the root cause of the downtime from earlier — about 3 hours, as best as I can tell. Seems it was a regression related to backend restructuring from a couple weeks ago. Sorry, this one was bad!

Two related fixes went out:

• Fixed a bug where real-time sync wasn’t working properly if a user was using the app on two devices.
• Fixed intermittent page load errors caused by note sync holding open too many database connections. Background tabs were keeping connections alive unnecessarily; now they release them when inactive.

03:15

• Switching subscription plans is now smoother — including a new option to switch to monthly billing
• Fixed a bug where your view mode (plain text vs. rendered) would reset after AI updated your tags
• Grug tidied up the cave drawings on the wall (changelog cleanup, internal bookkeeping)

04:59

• Grug scrubbed strange markings off the cave walls (internal cleanup, AI metadata sanitization)

09:09

• hmm some strange database connection issues the last 5 hours. Suspect a google issue, but investigating.

09:56 -- seems to be a google issue as far as I can tell, but monitoring.

11:48

• Prompts with non-English names (Japanese, Chinese, emoji, symbols, etc.) now work correctly — you can create and navigate to prompts in any language
• Fixed a bug where AI-written notes could land in the wrong space when saving
• Grug reinforced the cave walls and patched a few cracks (internal stability improvements, error handling)

13:56

• Fixed an issue where web research could fail if started without all parameters — it now correctly picks up context from the parent chat
• Grug tidied the cave shelves (changelog bookkeeping)

21:09 BIG DEPLOY!

• Reminders! This is kinda awesome! Grug now classifies auto tagged dates as reminders, events, or mentions in your notes automatically. They show up on your timeline with color-coded date tags. Reminders stay pinned until checked off. Events show on the day of the event in the reminder section (top). And mentions show below the blocks.

• Web research tool in chat. Ask Grug to look something up — it'll search the web (powered by Gemini grounding). This is really useful for agents as they can proactively look stuff up for you!

• Grug Chat upgraded to GPT-5.4 smarter Grug.

• Added warning before accidental navigation while recording audio.

• Fixed some funky flow ux upgrading of downgrading plans and switching between monthly/annual plans.

• Grug organized the bone pile and trained a night-watch dinosaur to guard the cave entrance (internal tooling, automated error monitoring, code cleanup, tons of misc bug fixes).

00:00

• Fixed calendar block editing, which was broken in a recent regression

02:47

• Fixed an error that could occur when using the 2nd Opinion dropdown
• Fixed a bug where usernames with mixed capitalization would see errors applying chat blocks to notes

15:43

• AI-highlighted text (underlines) now stays intact when you save a note, regression from feb 17th

17:14

• Fixed a regression error that could occur when applying all from a chat to notes, ie the notes wouldn't show up until a refresh.
• Fixed tasks so it more reliably shows the diff automatically

19:21

• Fixed List button disappearing on calendar view

05:28

• Telegram replies now stream in real-time instead of arriving all at once
• Landing page copy and layout improvements
• Fixed an issue where email-based agents could cause repeated errors if the agent was deleted
• Agent access now properly activates and deactivates with your subscription

06:11

• Telegram bot no longer shows raw reasoning tags while streaming responses

07:12

• When you send a note via Telegram that doesn't need a reply, Grug now reacts with an emoji instead of staying silent — so you know your message was received.

Massive updates deployed. Monitoring for regressions over the next few days. Docs and videos coming soon on the major features. Also a lot of infrastructure changes to set groundwork for further improvements.

Lists — New "List" prompt mode for managing ordered collections with a full table view:

• Inline cell editing, search, sort, filter, pagination, and column visibility
• Drag-to-reorder rows and columns
• Reference cells with typeahead linking between lists (stored as [[wikilinks]])
• Ref cell popovers for viewing/editing linked item attributes
• Boolean cells render as auto-saving checkboxes
• Adaptive wide layout for tables with many columns
• Lists section in the sidebar
• "Tracker" auto-detected as list mode for new prompts

Wikilinks & References — [[prompt/block]] wikilink support with automatic path sync when prompts are renamed. Block text references are rewritten on rename too.

Entity Merge — Merging duplicate entities now rewrites references across your notes and handles conflicts gracefully.

AI Backlinks — Incremental hash-based diffing replaces the old delete-all/recreate approach. Backlinks now have a confidence level that locks on confirm.

Sage Model — New feature release powered by Claude Sonnet 4.6. Ungated for all users.

Gemini — Upgraded to 3.1 Pro.

2nd Opinion — Model selection dropdown lets you pick which model to use.

Search — Switched off Gemini flash 2 models to Groq/Gemini Flash 3. Smarter results interpretation.

Date Extraction — Improved accuracy for relative weeks, appointment patterns, and HTML-escaped text.

Prompt Attributes — Drag-and-drop sorting, default values, and descriptions for prompt attributes.

Autotitle — Blocks can now be auto-titled.

Collapse/Expand — Toggle to collapse or expand all blocks on a page. List blocks default to collapsed.

URL Linkification — Fixed streamed and markdown-wrapped URLs not becoming clickable links.

Enter to Submit — Input-style submit controls now respond to Enter key.

Fixes

• Chat textarea intermittently hidden after focus
• Month-boundary date prefilter off-by-one
• Scrollbar flicker on prompt header
• Layout flash on page load
• Stale list table after browser back
• "Past" header missing when recent days are empty
• Collapse arrow unreachable on prompt/block pages
• Code block copy button hidden in some cases
• Blocks created under wrong prompt
• Stale static assets served from home cache after deploy
• CSRF error on ASR token count
• Duplicate loading spinners on Publish/Pin buttons
• Sidebar caches not refreshing after async emoji generation
• Stale flash messages on tasks page
• Focus prompt checkbox removed from prompt settings

Fireworks AI - Fastest Inference for Generative AI

Fetched from: https://fireworks.ai Date: 2026-01-28 07:37 Published: 2025-09-24

"Fireworks has been a fantastic partner in building AI dev tools at Sourcegraph. Their fast, reliable model inference lets us focus on fine-tuning, AI-powered code search, and deep code context, making Cody the best AI coding assistant. They are responsive and ship at an amazing pace."

"By partnering with Fireworks to fine-tune models, we reduced latency from about 2 seconds to 350 milliseconds, significantly improving performance and enabling us to launch AI features at scale. That improvement is a game changer for delivering reliable, enterprise-scale AI"

“Fireworks has been an amazing partner getting our Fast Apply and Copilot++ models running performantly. They exceeded other competitors we reviewed on performance. After testing their quantized model quality for our use cases, we have found minimal degradation. Fireworks helps implement task specific speed ups and new architectures, allowing us to achieve bleeding edge performance!”

"We've had a really great experience working with Fireworks to host open source models, including SDXL, Llama, and Mistral. After migrating one of our models, we noticed a 3x speedup in response time, which made our app feel much more responsive and boosted our engagement metrics."

"Fireworks has been a fantastic partner in building AI dev tools at Sourcegraph. Their fast, reliable model inference lets us focus on fine-tuning, AI-powered code search, and deep code context, making Cody the best AI coding assistant. They are responsive and ship at an amazing pace."

"By partnering with Fireworks to fine-tune models, we reduced latency from about 2 seconds to 350 milliseconds, significantly improving performance and enabling us to launch AI features at scale. That improvement is a game changer for delivering reliable, enterprise-scale AI"

“Fireworks has been an amazing partner getting our Fast Apply and Copilot++ models running performantly. They exceeded other competitors we reviewed on performance. After testing their quantized model quality for our use cases, we have found minimal degradation. Fireworks helps implement task specific speed ups and new architectures, allowing us to achieve bleeding edge performance!”

"We've had a really great experience working with Fireworks to host open source models, including SDXL, Llama, and Mistral. After migrating one of our models, we noticed a 3x speedup in response time, which made our app feel much more responsive and boosted our engagement metrics."

pretty massive deploy of new audio features.

• The meeting and lock buttons now allow the rest of the app to be usuable while recording (don't click links however).
• The meeting uses Eleven Labs speech to text now, vastly better results for multi-speaker audio
• There is a new recording page under menu -- recordings are stored realtime locally, so much less chance of data loss! This is a big step!
• Meeting and Lock recordings now get put in sub-blocks automatically...

Will me making some adjustments over the next few days, but this is a big step!