* Adds a fast / cheaper model to providers and defaults thread
summarization to this model. Initial motivation for this was that
https://github.com/zed-industries/zed/pull/29099 would cause these
requests to fail when used with a thinking model. It doesn't seem
correct to use a thinking model for summarization.
* Skips system prompt, context, and thinking segments.
* If tool use is happening, allows 2 tool uses + one more agent response
before summarizing.
Downside of this is that there was potential for some prefix cache reuse
before, especially for title summarization (thread summarization omitted
tool results and so would not share a prefix for those). This seems fine
as these requests should typically be fairly small. Even for full thread
summarization, skipping all tool use / context should greatly reduce the
token use.
Release Notes:
- N/A
Previously, we were including the dummy text "Here are the tool
results." whenever reporting tool call results. I'm worried this is
adding noise and confusing the model, because the user didn't actually
say anything. This inserts an empty message to be populated later. My
preference would be something less stateful, where tool results (or
batches of them requested simultaneously) would be sent to the model as
soon as they were ready, without bothering to do this message
association dance. But for now, this seems to work.
Release Notes:
- N/A
This is based on user feedback that the Agent Panel menu was only
linking to extensions as a way to add MCP servers while we also support
adding "custom" servers, too, which don't go through the extensions
flow.
Release Notes:
- N/A
Now that we've established a proper eval in tree, this PR is reboots of
our agent loop back to a set of minimal tools and simpler prompts. We
should aim to get this branch feeling subjectively competitive with
what's on main and then merge it, and build from there.
Let's invest in our eval and use it to drive better performance of the
agent loop. How you can help: Pick an example, and then make the outcome
faster or better. It's fine to even use your own subjective judgment, as
our evaluation criteria likely need tuning as well at this point. Focus
on making the agent work better in your own subjective experience first.
Let's focus on simple/practical improvements to make this thing work
better, then determine how we can craft our judgment criteria to lock
those improvements in.
Release Notes:
- N/A
---------
Co-authored-by: Max <max@zed.dev>
Co-authored-by: Antonio <antonio@zed.dev>
Co-authored-by: Agus <agus@zed.dev>
Co-authored-by: Richard <richard@zed.dev>
Co-authored-by: Max Brunsfeld <maxbrunsfeld@gmail.com>
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Michael Sloan <mgsloan@gmail.com>
This PR attaches the thread ID and the new prompt ID to telemetry events
for completions in the Agent panel.
Release Notes:
- N/A
---------
Co-authored-by: Mikayla Maki <mikayla.c.maki@gmail.com>
Related to #28490.
- Default prompts from the prompt library are now included as "user
rules" in the system prompt.
- Presence of these user rules is shown at the beginning of the thread
in the UI.
_ Now uses an `Entity<PromptStore>` instead of an `Arc<PromptStore>`.
Motivation for this is emitting a `PromptsUpdatedEvent`.
- Now disallows concurrent reloading of the system prompt. Before this
change it was possible for reloads to race.
Release Notes:
- agent: Added support for including default prompts from the Prompt
Library as "user rules" in the system prompt.
---------
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
This PR makes it so the thread summarization also reports the model
request usage, to prevent the case where the count would appear to jump
by 2 the next time a message was sent after summarization.
Release Notes:
- N/A
This PR adds a banner showing request usage in the Agent panel:
<img width="640" alt="Screenshot 2025-04-17 at 5 51 46 PM"
src="https://github.com/user-attachments/assets/e0eb036c-57c1-441c-bbab-7dab1c6e56d9"
/>
Only visible to users on the new billing.
Note to Joseph: Doesn't need to be cherry-picked to Preview.
Release Notes:
- N/A
---------
Co-authored-by: Nate <nate@zed.dev>
This PR updates the Agent to extract the usage information from the
response headers, if they are present.
For now we just log the information, but we'll be using this soon to
populate some UI.
Release Notes:
- N/A
Just now realized that the reason this button wasn't working reliably is
because we weren't passing the index to it. It's now fixed.
Release Notes:
- N/A
The UI was mistakenly using the cumulative token usage for the token
counter. It will now display the last request token count, plus an
estimation of the tokens in the message editor and context entries that
haven't been sent yet.
https://github.com/user-attachments/assets/0438c501-b850-4397-9135-57214ca3c07a
Additionally, when the user edits a message, we'll display the actual
token count up to it and estimate the tokens in the new message.
Note: We don't currently estimate the delta when switching profiles. In
the future, we want to use the count tokens API to measure every part of
the request and display a breakdown.
Release Notes:
- agent: Made the token count more accurate and added back estimation of
used tokens as you type and add context.
---------
Co-authored-by: Bennet Bo Fenner <bennetbo@gmx.de>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
This makes the `assistant: Quote selection` work again for text threads.
Next up is supporting this also in normal threads.
Release Notes:
- agent: Add support for inserting selections (assistant: Quote
selection) into text threads
Make fuzzy search in @-mention match paths and context kinds as well
(e.g., typing "sym" should let me select the "Symbols" label, as opposed
to just paths)
Release Notes:
- agent: Improve fuzzy-matching when using @mentions
Staff only for now. We'll work on making this usable for non zed.dev
users later
Release Notes:
- N/A
---------
Co-authored-by: Antonio Scandurra <me@as-cii.com>
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Marshall Bowers <git@maxdeviant.com>
This PR adds an error message when the model requests limit has been
hit.
Release Notes:
- N/A
Co-authored-by: Oleksiy Syvokon <oleksiy.syvokon@gmail.com>
If you wanted to interrupt the current LLM response that's generating to
send a follow up message, you'd need to stop it first, type your new
message, and then send it. Now, you can just type your new message while
there's a response generating and send it. This will interrupt the
previous response generation and kick off a new one.
Release Notes:
- agent: Allow to send a new message while a response is generating,
interrupting the LLM to focus instead on the most recent prompt.
This is just a refactor which adds no functionality.
We now return a `ToolResult` from `Tool > run(...)`. For now this just
wraps the output task in a struct. We'll use this to implement custom
rendering of tools, see #28621.
Release Notes:
- N/A
Ended up not making this one visible only upon hover or something
because the layout alignment would be weird given the list item spans
the full width. So, experimenting with this design here:
<img
src="https://github.com/user-attachments/assets/62bf661e-1aae-4644-8a89-49cefb3e8130"
width="700" />
Release Notes:
- agent: Add scrollbar to the history view.
When you hit the back button in the agent panel toolbar, we were
returning the focus to the buffer instead to the panel's message editor,
which is likely where you want to be after quickly checking history or
settings.
Release Notes:
- N/A
Allow expanding the message editor while the agent panel is not focused,
right now there is no effect when you use the button from another focus
Release Notes:
- N/A
This PR splits up the rendering of the message editor into multiple
functions. Previously we had a single `h_flex()...` expression which
spanned across 550 lines, `cargo fmt` stopped working.
Release Notes:
- N/A
Release Notes:
- agent: Show recommended models in the agent model selector and display
the provider in the model selector's trigger.
---------
Co-authored-by: Danilo Leal <daniloleal09@gmail.com>
Co-authored-by: Danilo Leal <67129314+danilo-leal@users.noreply.github.com>
Agus Zubiaga
Michael Sloan
Nathan Sobo
Danilo Leal
Marshall BowersandMikayla Maki
5brian