OpenCode CLI

← back to overview

Source of truth: OpenCode's docs for config keys; the overview for the gateway endpoint and context size. If this page disagrees with them, they win.

Confirmed project: anomalyco/opencode (renamed from sst/opencode — don't confuse with the unrelated opencode-ai/opencode Go TUI). Docs: https://opencode.ai/docs/

Install:

curl -fsSL https://opencode.ai/install | bash

Config (opencode.json, project root or ~/.config/opencode/opencode.json):

{
    "$schema": "https://opencode.ai/config.json",
    "model": "omniroute/qwen3.8-27b-local//models/Qwen3.8-27B-UD-Q4_K_XL.gguf",
    "provider": {
        "omniroute": {
            "npm": "@ai-sdk/openai-compatible",
            "name": "OmniRoute",
            "options": {
                "baseURL": "https://<YOUR-OMNIROUTE-PROXY>/v1"
            },
            "models": {
                "qwen3.8-27b-local//models/Qwen3.8-27B-UD-Q4_K_XL.gguf": {
                    "name": "Qwen3.8-27B",
                    "reasoning": true,
                    "interleaved": {
                        "field": "reasoning_content"
                    },
                    "limit": {
                        "context": 131000,
                        "output": 65536
                    },
                    "cost": {
                        "input": 0.4,
                        "output": 3.0,
                        "cache": {
                            "read": 0.04,
                            "write": 0
                        }
                    }
                }
            }
        }
    }
}

limit.context is the per-slot context (overview). OpenCode uses it for its own context bookkeeping; the server doesn't read it.

Select the model with llm-server/qwen3.8-27b-local.

OpenCode-specific risks (on top of the shared Qwen3.8-27B tool-calling risk — see overview):

Relies on the backend's --jinja flag (architecture → llama-server); without it, OpenCode's unconditional tool-call scaffolding gets a 500.
anomalyco/opencode#20669 (closed as "not planned" — a live, unfixed risk): OpenCode's bash tool crashes if the model omits the optional description field on a tool call; some local backends return finish_reason: tool_calls with an empty array, which can hang the agent loop instead of stopping cleanly.
Thinking-mode handling (options.reasoningEffort) is undocumented for models that emit inline <think> tags rather than a native reasoning API field — expect no effect from that config on this model; untested.

Further reading: research/opencode-cli-setup.md (its base URL example predates the gateway), research/opencode-auto-compact.md.

2.9 KiB Raw Blame History

OpenCode CLI

2.9 KiB

Raw Blame History