Github Copilot: Tracking Premium Requests #556

Open
opened 2026-02-16 17:27:18 -05:00 by yindo · 26 comments
Owner

Originally created by @FareedFarag on GitHub (Jul 8, 2025).

Originally assigned to: @thdxr on GitHub.

When using Github copilot models, the cost tracker is irrelevant as it always stays at $0.00. Instead, is it possible to display the premium requests quota? See below image as an example from VS code copilot extension.

Image
Originally created by @FareedFarag on GitHub (Jul 8, 2025). Originally assigned to: @thdxr on GitHub. When using Github copilot models, the cost tracker is irrelevant as it always stays at $0.00. Instead, is it possible to display the premium requests quota? See below image as an example from VS code copilot extension. <img width="241" height="177" alt="Image" src="https://github.com/user-attachments/assets/0c560261-90fc-4a67-90ab-88df7b6af534" />
Author
Owner

@martinhpedersen commented on GitHub (Jul 8, 2025):

FYI - Don't recall where I found it, but the data is available through this endpoint:

curl https://api.github.com/copilot_internal/user -H "Authorization: Bearer $COPILOT_API_KEY" -H "Content-Type: application/json"

If anyone is interested in a quick alias to get the current quota/usage:

alias copilot-quota="curl -s https://api.github.com/copilot_internal/user -H 'Authorization: Bearer $COPILOT_API_KEY' -H 'Content-Type: application/json' | jq .quota_snapshots.premium_interactions"
@martinhpedersen commented on GitHub (Jul 8, 2025): FYI - Don't recall where I found it, but the data is available through this endpoint: `curl https://api.github.com/copilot_internal/user -H "Authorization: Bearer $COPILOT_API_KEY" -H "Content-Type: application/json"` If anyone is interested in a quick alias to get the current quota/usage: ```sh alias copilot-quota="curl -s https://api.github.com/copilot_internal/user -H 'Authorization: Bearer $COPILOT_API_KEY' -H 'Content-Type: application/json' | jq .quota_snapshots.premium_interactions" ```
Author
Owner

@FareedFarag commented on GitHub (Jul 9, 2025):

FYI - Don't recall where I found it, but the data is available through this endpoint:

curl https://api.github.com/copilot_internal/user -H "Authorization: Bearer $COPILOT_API_KEY" -H "Content-Type: application/json"

If anyone is interested in a quick alias to get the current quota/usage:

alias copilot-quota="curl -s https://api.github.com/copilot_internal/user -H 'Authorization: Bearer $COPILOT_API_KEY' -H 'Content-Type: application/json' | jq .quota_snapshots.premium_interactions"

How do you get COPILOT_API_KEY? Copilot doesn't issue API keys as far as I know.

@FareedFarag commented on GitHub (Jul 9, 2025): > FYI - Don't recall where I found it, but the data is available through this endpoint: > > `curl https://api.github.com/copilot_internal/user -H "Authorization: Bearer $COPILOT_API_KEY" -H "Content-Type: application/json"` > > If anyone is interested in a quick alias to get the current quota/usage: > > alias copilot-quota="curl -s https://api.github.com/copilot_internal/user -H 'Authorization: Bearer $COPILOT_API_KEY' -H 'Content-Type: application/json' | jq .quota_snapshots.premium_interactions" How do you get `COPILOT_API_KEY`? Copilot doesn't issue API keys as far as I know.
Author
Owner

@asadmoosvi commented on GitHub (Jul 26, 2025):

As a github copilot user, I think this would be a great addition to the software. Not having to open up the browser to look at remaining premium requests would be such a good quality-of-life improvement.

@asadmoosvi commented on GitHub (Jul 26, 2025): As a github copilot user, I think this would be a great addition to the software. Not having to open up the browser to look at remaining premium requests would be such a good quality-of-life improvement.
Author
Owner

@berenar commented on GitHub (Oct 27, 2025):

Here's a curl that uses your GitHub Copilot Oauth token instead:

curl -s -H "Authorization: Bearer $(cat ~/.config/github-copilot/apps.json \
| jq '.[].oauth_token' \
| sed 's/^"\(.*\)"$/\1/')" https://api.github.com/copilot_internal/user

As a user of Copilot, having this info in the UI would be very useful. I'm happy to implement this if the authors also see its value.
cc @thdxr

@berenar commented on GitHub (Oct 27, 2025): Here's a curl that uses your GitHub Copilot Oauth token instead: ``` curl -s -H "Authorization: Bearer $(cat ~/.config/github-copilot/apps.json \ | jq '.[].oauth_token' \ | sed 's/^"\(.*\)"$/\1/')" https://api.github.com/copilot_internal/user ``` As a user of Copilot, having this info in the UI would be very useful. I'm happy to implement this if the authors also see its value. cc @thdxr
Author
Owner

@rekram1-node commented on GitHub (Oct 27, 2025):

@berenar how would you go about implementing it? There will probably be other providers that have special billing things like copilot too

This would be welcome but we would probably need to discuss how it would be shown in the ui

@rekram1-node commented on GitHub (Oct 27, 2025): @berenar how would you go about implementing it? There will probably be other providers that have special billing things like copilot too This would be welcome but we would probably need to discuss how it would be shown in the ui
Author
Owner

@berenar commented on GitHub (Oct 28, 2025):

Providers support

Well, so far I've only used Claude (subscription based), GitHub Copilot and Zen.

  • Anthropic has hourly and daily usage limits. → And an API to get that info
  • Copilot "has X premium requests per month" → And an (internal) API endpoint to get that info too

I think the rest of supported providers are not subscription based (yet...), therefore they would not need this feature

UI

On how to show it in the UI, that's yours to decide :)

I would leave the top right corner for info related to the current session

  • Context size
  • Context %
  • Cost so far
    Image

And instead use the bottom right corner to show the provider limits
Image

The info to display would change based on the provider being used

Claude

  • "Approaching hourly limit" Same as how Claude Code shows it
  • "Approaching daily limit"
  • "Limit resets at 3pm"

Copilot

  • "80% used Premium Requests. Resets Nov. 1st"

wdyt @rekram1-node ?

@berenar commented on GitHub (Oct 28, 2025): # Providers support Well, so far I've only used Claude (subscription based), GitHub Copilot and Zen. - Anthropic has hourly and daily usage limits. → And an [API](https://docs.claude.com/en/api/usage-cost-api) to get that info - Copilot "has X [premium requests](https://docs.github.com/en/copilot/concepts/billing/copilot-requests#model-multipliers) per month" → And an (internal) [API endpoint](https://github.com/sst/opencode/issues/768#issuecomment-3450011296) to get that info too I think the rest of supported providers are not subscription based (yet...), therefore they would not need this feature # UI On how to show it in the UI, that's yours to decide :) I would leave the top right corner for info related to the current session - Context size - Context % - Cost so far <img width="357" height="244" alt="Image" src="https://github.com/user-attachments/assets/b5d42d6d-7fe4-4a71-aa69-e4ef1be77c2e" /> And instead use the bottom right corner to show the provider limits <img width="407" height="193" alt="Image" src="https://github.com/user-attachments/assets/092f9211-99a0-4814-8863-6315ce075d78" /> The info to display would change based on the provider being used ## Claude - "Approaching hourly limit" Same as how Claude Code [shows it](https://www.reddit.com/r/ClaudeCode/comments/1o06m96/what_are_these_claude_code_limits/) - "Approaching daily limit" - "Limit resets at 3pm" ## Copilot - "80% used Premium Requests. Resets Nov. 1st" wdyt @rekram1-node ?
Author
Owner

@rekram1-node commented on GitHub (Oct 29, 2025):

Sorry for delay @berenar I think we should wait for opentui to be wrapped up before a PR is attempted here. Another thing we need to surface that currently isn't being surfaced is retries.

I recently added some logic that will allow us to display the # of retries, if it is retrying, etc.

And that would also tie into this because some plan quotas are rate limited and we do exponential backoffs with awareness of the rate limit headers so there are a few moving pieces here.

That said I do like your proposal I just thing it will be best to revisit in a week once opentui has a little less stuff in motion

@rekram1-node commented on GitHub (Oct 29, 2025): Sorry for delay @berenar I think we should wait for opentui to be wrapped up before a PR is attempted here. Another thing we need to surface that currently isn't being surfaced is retries. I recently added some logic that will allow us to display the # of retries, if it is retrying, etc. And that would also tie into this because some plan quotas are rate limited and we do exponential backoffs with awareness of the rate limit headers so there are a few moving pieces here. That said I do like your proposal I just thing it will be best to revisit in a week once opentui has a little less stuff in motion
Author
Owner

@itse4elhaam commented on GitHub (Nov 25, 2025):

Hey, would love this!!!

Any plans on this?

@itse4elhaam commented on GitHub (Nov 25, 2025): Hey, would love this!!! Any plans on this?
Author
Owner

@rekram1-node commented on GitHub (Nov 26, 2025):

No immediate plans but we should think about how we wanna show it

@rekram1-node commented on GitHub (Nov 26, 2025): No immediate plans but we should think about how we wanna show it
Author
Owner

@flupkede commented on GitHub (Dec 4, 2025):

This same issue applies to z.ai Coding Plans. Cost tracking shows $0.00, but users have no visibility into their prompt quota. Even worse, z.ai does not even have an api call to track this.

Not ideal however a possible approach:

  • Optional quotas config per provider, make it configurable per model: for example glm4.6: Lite: 120 prompts/5h , Pro: 600 prompts/5h, Max: 2400 prompts/5h
  • Track rolling window prompt count per configurable time window.
  • Display alongside cost/even hide cost for flat fee plan: "Cost: $0.00 | Usage: 87%" example based upon 104 requests of 120 limit)
@flupkede commented on GitHub (Dec 4, 2025): This same issue applies to z.ai Coding Plans. Cost tracking shows $0.00, but users have no visibility into their prompt quota. Even worse, z.ai does not even have an api call to track this. Not ideal however a possible approach: - Optional `quotas` config per provider, make it configurable per model: for example glm4.6: Lite: 120 prompts/5h , Pro: 600 prompts/5h, Max: 2400 prompts/5h - Track rolling window prompt count per configurable time window. - Display alongside cost/even hide cost for flat fee plan: "Cost: $0.00 | Usage: 87%" example based upon 104 requests of 120 limit)
Author
Owner

@ojsef39 commented on GitHub (Dec 11, 2025):

Would love to see this for Github and Anthropic, i use both and id love to see how much time i have left before i have to think for myself again :D

@ojsef39 commented on GitHub (Dec 11, 2025): Would love to see this for Github and Anthropic, i use both and id love to see how much time i have left before i have to think for myself again :D
Author
Owner

@rosematcha commented on GitHub (Dec 17, 2025):

Agreed, would love to see this implemented for Copilot, Anthropic, and Codex subscription plans!

@rosematcha commented on GitHub (Dec 17, 2025): Agreed, would love to see this implemented for Copilot, Anthropic, and Codex subscription plans!
Author
Owner

@peixotorms commented on GitHub (Dec 29, 2025):

Copilot seems straightforward with just an api call, please add it to opencode.

@peixotorms commented on GitHub (Dec 29, 2025): Copilot seems straightforward with just an api call, please add it to opencode.
Author
Owner

@amaduswaray commented on GitHub (Jan 7, 2026):

We just started using OpenCode at work, and this would be useful! Are there any requirements beyond the contribution guide to submit a PR on this?

@amaduswaray commented on GitHub (Jan 7, 2026): We just started using OpenCode at work, and this would be useful! Are there any requirements beyond the contribution guide to submit a PR on this?
Author
Owner

@mathisdev7 commented on GitHub (Jan 8, 2026):

i ran into the same issue and ended up building a small focused fork called arctic that only cares about plan-based quotas and limits.
if quota visibility is the main pain, this might be useful: https://github.com/arctic-cli/interface

@mathisdev7 commented on GitHub (Jan 8, 2026): i ran into the same issue and ended up building a small focused fork called arctic that only cares about plan-based quotas and limits. if quota visibility is the main pain, this might be useful: https://github.com/arctic-cli/interface
Author
Owner

@heliar-k commented on GitHub (Jan 17, 2026):

I also need these, as well as daily quota tracking for OpenAI Codex.

@heliar-k commented on GitHub (Jan 17, 2026): I also need these, as well as daily quota tracking for OpenAI Codex.
Author
Owner

@bew commented on GitHub (Jan 20, 2026):

There should also be a way to show the model multiplier in the models selection UI & in the upcoming usage UI / sidebar

Would also be nice to sort the list of models by that multiplier 🤔

@bew commented on GitHub (Jan 20, 2026): There should also be a way to show [the model multiplier](https://docs.github.com/en/copilot/concepts/billing/copilot-requests#model-multipliers) in the models selection UI & in the upcoming usage UI / sidebar Would also be nice to sort the list of models by that multiplier 🤔
Author
Owner

@sTiKyt commented on GitHub (Jan 30, 2026):

@berenar how would you go about implementing it? There will probably be other providers that have special billing things like copilot too

This would be welcome but we would probably need to discuss how it would be shown in the ui

Why not somewhere around here?

Image
@sTiKyt commented on GitHub (Jan 30, 2026): > [@berenar](https://github.com/berenar) how would you go about implementing it? There will probably be other providers that have special billing things like copilot too > > This would be welcome but we would probably need to discuss how it would be shown in the ui Why not somewhere around here? <img width="379" height="483" alt="Image" src="https://github.com/user-attachments/assets/b7b5051a-2212-4b8a-8aa0-8235178b7d88" />
Author
Owner

@heliar-k commented on GitHub (Jan 31, 2026):

Any Progress of this?

@heliar-k commented on GitHub (Jan 31, 2026): Any Progress of this?
Author
Owner

@Joost1982 commented on GitHub (Feb 6, 2026):

There should also be a way to show the model multiplier in the models selection UI & in the upcoming usage UI / sidebar

I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot /models API only returns billing.multiplier if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation.

What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read billing.multiplier from /models. See: https://github.com/Joost1982/opencode/commit/4c2cf2a348bddf24e7c43fda623a313fb6a2473f
It works on my machine (Debian), but it’s no real solution.

Image
@Joost1982 commented on GitHub (Feb 6, 2026): > There should also be a way to show [the model multiplier](https://docs.github.com/en/copilot/concepts/billing/copilot-requests#model-multipliers) in the models selection UI & in the upcoming usage UI / sidebar I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot **/models** API only returns **billing.multiplier** if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation. What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read **billing.multiplier** from **/models**. See: https://github.com/Joost1982/opencode/commit/4c2cf2a348bddf24e7c43fda623a313fb6a2473f It works on my machine (Debian), but it’s no real solution. <img width="503" height="565" alt="Image" src="https://github.com/user-attachments/assets/6e47db94-7692-4766-9a7e-a906822c0001" />
Author
Owner

@Joost1982 commented on GitHub (Feb 6, 2026):

Premium requests info can be added like this: https://github.com/Joost1982/opencode/commit/f217b38ba6bb1f038dc72272718ded18e11b7961
Only shows info when having a GitHub Copilot model selected:

Image

(uses the github token known by opencode, not the vsc / github cli token like in the comment above)

@Joost1982 commented on GitHub (Feb 6, 2026): Premium requests info can be added like this: https://github.com/Joost1982/opencode/commit/f217b38ba6bb1f038dc72272718ded18e11b7961 Only shows info when having a GitHub Copilot model selected: <img width="1008" height="934" alt="Image" src="https://github.com/user-attachments/assets/6ec31a61-3497-416f-8c18-e5912278077d" /> (uses the github token known by opencode, not the vsc / github cli token like in the comment above)
Author
Owner

@berenar commented on GitHub (Feb 6, 2026):

There should also be a way to show the model multiplier in the models selection UI & in the upcoming usage UI / sidebar

I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot /models API only returns billing.multiplier if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation.

What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read billing.multiplier from /models. See: https://github.com/Joost1982/opencode/commit/4c2cf2a348bddf24e7c43fda623a313fb6a2473f
It works on my machine (Debian), but it’s no real solution.

Image

What about just hardcoding the multipliers? They rarely change

@berenar commented on GitHub (Feb 6, 2026): > > There should also be a way to show [the model multiplier](https://docs.github.com/en/copilot/concepts/billing/copilot-requests#model-multipliers) in the models selection UI & in the upcoming usage UI / sidebar > > I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot **/models** API only returns **billing.multiplier** if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation. > > What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read **billing.multiplier** from **/models**. See: https://github.com/Joost1982/opencode/commit/4c2cf2a348bddf24e7c43fda623a313fb6a2473f > It works on my machine (Debian), but it’s no real solution. > > <img width="503" height="565" alt="Image" src="https://github.com/user-attachments/assets/6e47db94-7692-4766-9a7e-a906822c0001" /> What about just hardcoding the multipliers? They rarely change
Author
Owner

@slhad commented on GitHub (Feb 6, 2026):

I have seen Claude Opus 4.5 (preview) gone from 1x to 3x when "preview" title was removed, and it made me cry when I found out I was not hallucinating in VS Code,

Theses changes can be devastating to the premium requests count (been here), I would be nice if it does not rely on manual update of hardcoded constants

@slhad commented on GitHub (Feb 6, 2026): I have seen Claude Opus 4.5 (preview) gone from 1x to 3x when "preview" title was removed, and it made me cry when I found out I was not hallucinating in VS Code, Theses changes can be devastating to the premium requests count (been here), I would be nice if it does not rely on manual update of hardcoded constants
Author
Owner

@sTiKyt commented on GitHub (Feb 11, 2026):

Premium requests info can be added like this: Joost1982@f217b38 Only shows info when having a GitHub Copilot model selected:

Image (uses the github token known by opencode, not the vsc / github cli token like in the comment above)

but would be nice to see numeric values, not percentages imo

@sTiKyt commented on GitHub (Feb 11, 2026): > Premium requests info can be added like this: [Joost1982@f217b38](https://github.com/Joost1982/opencode/commit/f217b38ba6bb1f038dc72272718ded18e11b7961) Only shows info when having a GitHub Copilot model selected: > > <img alt="Image" width="1008" height="934" src="https://private-user-images.githubusercontent.com/38584259/546275032-6ec31a61-3497-416f-8c18-e5912278077d.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.QPDvGJQmNHzxMFp5CyA-nxPRF65coB82KWqmqlRnurY"> > (uses the github token known by opencode, not the vsc / github cli token like in the comment above) but would be nice to see numeric values, not percentages imo
Author
Owner

@sTiKyt commented on GitHub (Feb 11, 2026):

There should also be a way to show the model multiplier in the models selection UI & in the upcoming usage UI / sidebar

I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot /models API only returns billing.multiplier if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation.
What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read billing.multiplier from /models. See: Joost1982@4c2cf2a
It works on my machine (Debian), but it’s no real solution.
Image

What about just hardcoding the multipliers? They rarely change

they often change, sometimes they even get more expensive under high load/demand, and show stuff like this:

Image
@sTiKyt commented on GitHub (Feb 11, 2026): > > > There should also be a way to show [the model multiplier](https://docs.github.com/en/copilot/concepts/billing/copilot-requests#model-multipliers) in the models selection UI & in the upcoming usage UI / sidebar > > > > > > I dug into the multiplier adding to opencode, and it’s not straightforward. The Copilot **/models** API only returns **billing.multiplier** if you call it like an IDE (Editor-Version, Editor-Plugin-Version, Copilot-Integration-Id, VScode-SessionId/MachineId, plus a Copilot token). With the normal GitHub OAuth token opencode uses, the endpoint 404s or says “token not authorized,” so there’s no clean/official way to fetch multipliers with the current implementation. > > What I ended up doing is a bit hacky: I reuse a local VS Code Copilot app token to mint a Copilot token, send the IDE headers, and then read **billing.multiplier** from **/models**. See: [Joost1982@4c2cf2a](https://github.com/Joost1982/opencode/commit/4c2cf2a348bddf24e7c43fda623a313fb6a2473f) > > It works on my machine (Debian), but it’s no real solution. > > <img alt="Image" width="503" height="565" src="https://private-user-images.githubusercontent.com/38584259/546162999-6e47db94-7692-4766-9a7e-a906822c0001.png?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.hov7ya8yWJLl7NUD0Aq9kuBns0c7TR8rv9RyclG-am8"> > > What about just hardcoding the multipliers? They rarely change they often change, sometimes they even get more expensive under high load/demand, and show stuff like this: <img width="304" height="761" alt="Image" src="https://github.com/user-attachments/assets/d4d05e34-8fca-4056-bb75-252f846ec581" />
Author
Owner

@krzyk commented on GitHub (Feb 11, 2026):

they often change, sometimes they even get more expensive under high load/demand, and show stuff like this

How often? I did see just one change in 6 months - promotional 1x for Opus 4.5 for (1 or 2 weeks) and it was advertised that it will change. And similar for Grok Code, but somehow I still have it at 0x, while on copilot page it is listed as 0.25x.

I have not seen changes to multiplier based on load, your screenshot also doesn't show that. Models might be slow to respond, or not respond at all, but this is not change of multipliers.

@krzyk commented on GitHub (Feb 11, 2026): > they often change, sometimes they even get more expensive under high load/demand, and show stuff like this How often? I did see just one change in 6 months - promotional 1x for Opus 4.5 for (1 or 2 weeks) and it was advertised that it will change. And similar for Grok Code, but somehow I still have it at 0x, while on copilot page it is listed as 0.25x. I have not seen changes to multiplier based on load, your screenshot also doesn't show that. Models might be slow to respond, or not respond at all, but this is not change of multipliers.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#556