Skip to content

NVIDIA API models hang indefinitely with "Aborted process" - all models affected (DeepSeek V4, Llama, Mistral, Minimax) #24516

Closed as not planned
Closed as not planned
NVIDIA API models hang indefinitely with "Aborted process" - all models affected (DeepSeek V4, Llama, Mistral, Minimax)#24516
Assignees
nexxeln

Description

@alexbayov
alexbayov
opened on Apr 26, 2026

Description

Experiencing the same issue with NVIDIA API models via https://integrate.api.nvidia.com/v1

Configuration: All NVIDIA models (deepseek-ai/deepseek-v4-pro, meta/llama-3.3-70b-instruct, mistralai/mistral-large, minimaxai/minimax-m2.7) hang with "Размышляю" status and eventually fail with "Aborted process" error.

Environment:

  • OpenCode Version: 25
  • OS: macOS 26.2 (25C56)

Logs:
ERROR 2026-04-26T16:45:20 +230498ms service=session.processor error=Aborted process

Direct curl requests to NVIDIA API work correctly, suggesting streaming response handling issue.

Plugins

No response

OpenCode version

No response

Steps to reproduce

No response

Screenshot and/or share link

No response

Operating System

No response

Terminal

No response

Activity

  1. alexbayov
    added
    bugSomething isn't working
    on Apr 26, 2026
  2. github-actions
    assigned
    rekram1-node
    on Apr 26, 2026
  3. github-actions
    added
    coreAnything pertaining to core functionality of the application (opencode server stuff)
    on Apr 26, 2026

  4. github-actions commented on Apr 26, 2026

    @github-actions
    github-actionsBot
    on Apr 26, 2026 – with GitHub Actions
    Contributor

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

    • Nvidia NIM API hangs for DeepSeek v4 reasoning models without chat_template_kwargs #24264: Nvidia NIM API hangs for DeepSeek v4 reasoning models — same provider (NVIDIA NIM), same models (DeepSeek V4), and same symptom (hangs indefinitely). That issue identifies a specific root cause: NVIDIA NIM requires chat_template_kwargs in the payload for DeepSeek v4 reasoning models, which OpenCode currently doesn't inject. Your non-DeepSeek models (Llama, Mistral, Minimax) hanging may be a separate issue, but the DeepSeek V4 hang is likely the same.
  5. thdxr
    removed
    bugSomething isn't working
    coreAnything pertaining to core functionality of the application (opencode server stuff)
    on May 3, 2026
  6. thdxr
    assigned
    nexxeln
    and unassigned
    rekram1-node
    on May 3, 2026

  7. github-actions commented on Jul 4, 2026

    @github-actions
    github-actionsBot
    on Jul 4, 2026 – with GitHub Actions
    Contributor

    To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.

  8. github-actions
    closed this as not plannedon Jul 4, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions