get response blocked while using model deepseek-v4.1-flash. #57317

Activity

1. opencode-agent commented on Oct 20, 2026

   

   opencode-agentBot

   on Oct 20, 2026 – with OpenCode Agent

   Contributor

   More actions

   Thanks for the report! We can't read everything in the screenshot, so a few details would help us work out what's going on:

   • The exact error text, pasted as text. If it says "Provider blocked the response", please include anything after it, such as a category or explanation. That message means the model provider refused the request on content grounds. It isn't usually caused by token count. For example, in Provider blocked the response #57251 credentials in the conversation triggered it.
   • Provider: OpenCode Go/Zen, DeepSeek's own API, or something else?
   • Input tokens: roughly how many input tokens you're seeing, how far into the session it happens, and whether it also happens in a brand-new session with a short prompt.
   • The plugin: does it still happen with the OMO-Slim plugin disabled? Plugins can add a lot of context to every request.
   • What you were doing when it happened (roughly the prompt or task), and your operating system.

   You can reply here with the details, or edit the issue. If you edit it, please also leave a comment saying you've updated it, so we know you're done editing.

   These issues might be related, ordered from most to least likely:

   • Provider blocked the response #57251: "Provider blocked the response" from OpenCode. There it was caused by sensitive content (credentials) in the conversation.
   • DeepSeek v4.1 Flash always returns 400 #54117: DeepSeek v4.1 Flash always returns 400.
   • OpenCode Go: DeepSeek V4.1 Flash returns HTTP 500 while GLM works with the same credentials #56721: OpenCode Go DeepSeek V4.1 Flash returns HTTP 500.
   • compaction: auto-compaction loops when provider reports input tokens above the model's context limit #50474 / keep.tokens is not honoured: the compaction split walk-back is unbounded and skips synthetic messages, so agent-driven sessions carry 234K past a 15K setting #43250: input tokens growing larger than expected, in auto-compaction loops and with keep.tokens not being honoured.
2. opencode-agent

   added

   triaging

   on Oct 20, 2026
3. wenhuiwang93 commented on Oct 20, 2026

   

   wenhuiwang93

   on Oct 20, 2026

   Author

   More actions

   • Provider blocked the response

   • yes, error with Provider blocked the response.
   • Provider: OpenCode Go
   • Input tokens: 626930
4. opencode-agent

   closed this as not plannedon Oct 23, 2026
5. opencode-agent commented on Oct 23, 2026

   

   opencode-agentBot

   on Oct 23, 2026 – with OpenCode Agent

   Contributor

   More actions

   We haven't heard back, so we're closing this issue for now. Without the exact error text, the provider, and whether it still happens with OMO-Slim disabled, we can't tell whether this was the provider refusing the request on content grounds or a bug in OpenCode.

   If it happens again, please reopen this issue or open a new one with those details. If the error was "Provider blocked the response", #57251 may explain it: there, sensitive content in the conversation triggered the block. Thanks!
6. opencodejr

   reopened this on Nov 3, 2026
7. neriousy

   closed this as not plannedon Nov 12, 2026