anomalyco/opencode
1
0
Fork 0
mirror of https://github.com/anomalyco/opencode.git synced 2026-07-22 02:05:26 -04:00
Code Issues 5.9k Packages Projects Releases 19 Wiki Activity

GLM 4.7 #4744

New Issue
Closed
opened 2026-02-16 17:45:14 -05:00 by yindo · 13 comments
yindo commented 2026-02-16 17:45:14 -05:00
Owner
Copy Link
Copy Source

Originally created by @PRIMExALBIN on GitHub (Jan 11, 2026).

Originally assigned to: @adamdotdevin on GitHub.

Description

GLM 4.7 is extremely slow it took 2 minutes to respond to a Hi

Plugins

Oh My opencode

OpenCode version

im in the desktop version

Steps to reproduce

No response

Screenshot and/or share link

No response

Operating System

Windows 11

Terminal

Windows Terminal

Originally created by @PRIMExALBIN on GitHub (Jan 11, 2026). Originally assigned to: @adamdotdevin on GitHub. ### Description GLM 4.7 is extremely slow it took 2 minutes to respond to a Hi ### Plugins Oh My opencode ### OpenCode version im in the desktop version ### Steps to reproduce _No response_ ### Screenshot and/or share link _No response_ ### Operating System Windows 11 ### Terminal Windows Terminal
yindo added the windowsbugperfweb labels 2026-02-16 17:45:14 -05:00
yindo closed this issue 2026-02-16 17:45:14 -05:00
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@github-actions[bot] commented on GitHub (Jan 11, 2026):

This issue might be a duplicate of existing issues. Please check:

  • #7692: [Bug] JSON Parse Error with Zhipu GLM-4.7: Stream chunks are concatenated incorrectly - may cause response delays
  • #7779: GLM 4.7 thinking process not properly formatted - missing leading tags - could affect performance
  • #6074: Differences between the free and paid versions of GLM 4.7 - discusses response speed differences

Feel free to ignore if none of these address your specific case.

@github-actions[bot] commented on GitHub (Jan 11, 2026): This issue might be a duplicate of existing issues. Please check: - #7692: [Bug] JSON Parse Error with Zhipu GLM-4.7: Stream chunks are concatenated incorrectly - may cause response delays - #7779: GLM 4.7 thinking process not properly formatted - missing leading tags - could affect performance - #6074: Differences between the free and paid versions of GLM 4.7 - discusses response speed differences Feel free to ignore if none of these address your specific case.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@Nsomnia commented on GitHub (Jan 11, 2026):

Because it's overloaded. Are you using any plugina of note? Oh oh-my-opencode this is typical and another model should be configured.

@Nsomnia commented on GitHub (Jan 11, 2026): Because it's overloaded. Are you using any plugina of note? Oh oh-my-opencode this is typical and another model should be configured.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@flupkede commented on GitHub (Jan 11, 2026):

Make sure you limit the context to max 100K.

@flupkede commented on GitHub (Jan 11, 2026): Make sure you limit the context to max 100K.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@RealLukeManning commented on GitHub (Jan 11, 2026):

Hitting the same. Random inexplicable delays. Lots of HTTP timeouts.

Getting 1303 errors very frequently: https://docs.z.ai/api-reference/api-code
High frequency usage of this API, please reduce frequency or contact customer service to increase limits

I have no plugins. I am getting this with very little context usage.

Using glm-4.7-free

@RealLukeManning commented on GitHub (Jan 11, 2026): Hitting the same. Random inexplicable delays. Lots of HTTP timeouts. Getting 1303 errors very frequently: https://docs.z.ai/api-reference/api-code `High frequency usage of this API, please reduce frequency or contact customer service to increase limits` I have no plugins. I am getting this with very little context usage. Using **glm-4.7-free**
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@lezhnev74 commented on GitHub (Jan 12, 2026):

I am using z.ai Coding Plan with GLM-4.7 and kind of frustrated of the speed. I guess it is ok for the price I pay, but compared to snappy opus 4.5 (which begun to shrink their quotas agressively and everybody is mad) it is unbearable.

@lezhnev74 commented on GitHub (Jan 12, 2026): I am using z.ai Coding Plan with GLM-4.7 and kind of frustrated of the speed. I guess it is ok for the price I pay, but compared to snappy opus 4.5 (which begun to shrink their quotas agressively and everybody is mad) it is unbearable.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@flupkede commented on GitHub (Jan 12, 2026):

@lezhnev74 there must be something wrong, latency or context overload, because I have quite good experience with the model, it's pretty comparable with sonnet 4.5 in terms of speed, a bit less capable. The only thing I had was occasional looping where the model keeps on spitting out the same chinese character. I also had my terminal screwed up by chinese unicode characters, however by setting a codepage on the terminal this got solved as well.

@flupkede commented on GitHub (Jan 12, 2026): @lezhnev74 there must be something wrong, latency or context overload, because I have quite good experience with the model, it's pretty comparable with sonnet 4.5 in terms of speed, a bit less capable. The only thing I had was occasional looping where the model keeps on spitting out the same chinese character. I also had my terminal screwed up by chinese unicode characters, however by setting a codepage on the terminal this got solved as well.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@rekram1-node commented on GitHub (Jan 12, 2026):

this isnt really something on our end, providers get hammered

@rekram1-node commented on GitHub (Jan 12, 2026): this isnt really something on our end, providers get hammered
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@kadaluarsa commented on GitHub (Jan 21, 2026):

I am using z.ai Coding Plan with GLM-4.7 and kind of frustrated of the speed. I guess it is ok for the price I pay, but compared to snappy opus 4.5 (which begun to shrink their quotas agressively and everybody is mad) it is unbearable.

IMO better pay cerebras inference for GLM 4.7

@kadaluarsa commented on GitHub (Jan 21, 2026): > I am using z.ai Coding Plan with GLM-4.7 and kind of frustrated of the speed. I guess it is ok for the price I pay, but compared to snappy opus 4.5 (which begun to shrink their quotas agressively and everybody is mad) it is unbearable. IMO better pay cerebras inference for GLM 4.7
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@mykola-dev commented on GitHub (Jan 25, 2026):

i'm on z.ai pro subscription and can barely cross the 5% of 5 hours quota. this is ridiculous.

@mykola-dev commented on GitHub (Jan 25, 2026): i'm on z.ai pro subscription and can barely cross the 5% of 5 hours quota. this is ridiculous.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@mykola-dev commented on GitHub (Jan 26, 2026):

yes. it is definitely opencode issue. i tried to switch to claude code and the GLM is now like a rocket. like 3-5 times faster. and now i can reach 15-20% of 5 hours limit. it seems like opencode still uses lite free plan despite i connected with a pro plan.

@mykola-dev commented on GitHub (Jan 26, 2026): yes. it is definitely opencode issue. i tried to switch to claude code and the GLM is now like a rocket. like 3-5 times faster. and now i can reach 15-20% of 5 hours limit. it seems like opencode still uses lite free plan despite i connected with a pro plan.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@ferrao commented on GitHub (Jan 26, 2026):

yes. it is definitely opencode issue. i tried to switch to claude code and the GLM is now like a rocket. like 3-5 times faster. and now i can reach 15-20% of 5 hours limit. it seems like opencode still uses lite free plan despite i connected with a pro plan.

How do you know opencode is still using the lite free plan @mykola-dev ? I am also on the pro plan and experiencing incredibly slow responses. True that I work with a context of close to 100k tokens, but sometimes it can take like an hour to get the model to do what I have asked.

@ferrao commented on GitHub (Jan 26, 2026): > yes. it is definitely opencode issue. i tried to switch to claude code and the GLM is now like a rocket. like 3-5 times faster. and now i can reach 15-20% of 5 hours limit. it seems like opencode still uses lite free plan despite i connected with a pro plan. How do you know opencode is still using the lite free plan @mykola-dev ? I am also on the pro plan and experiencing incredibly slow responses. True that I work with a context of close to 100k tokens, but sometimes it can take like an hour to get the model to do what I have asked.
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@moonray commented on GitHub (Jan 26, 2026):

GLM 4.7 from z.ai with a max sub is a lot slower and seems "dumber" than the opencode/glm-4.7-free was
Why the discrepancy, and how can it be made to match opencode/glm-4.7-free?

@moonray commented on GitHub (Jan 26, 2026): GLM 4.7 from z.ai with a max sub is a lot slower and seems "dumber" than the opencode/glm-4.7-free was Why the discrepancy, and how can it be made to match opencode/glm-4.7-free?
yindo commented 2026-02-16 17:45:15 -05:00
Author
Owner
Copy Link
Copy Source

@jonathanm-tkf commented on GitHub (Jan 26, 2026):

It's the same for me is ridiculous 8 seconds to responde simple "Hi"? any suggestion for this problem

@jonathanm-tkf commented on GitHub (Jan 26, 2026): It's the same for me is ridiculous 8 seconds to responde simple "Hi"? any suggestion for this problem
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
acp
bad
bug
discussion
docs
documentation
enhancement
extension
good first issue
growth
help-wanted
mcp
model-problem
needs:compliance
nix
opentui
perf
pull-request
Mirrored from GitHub Pull Request
 tui
web
windows
zen
No Label bug perf web windows
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#4744
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?