#could you share your config for the

1 messages · Page 1 of 1 (latest)

queen dew
#

"models": {
"mode": "merge",
"providers": {
"local": {
"baseUrl": "http://xxx.xxx.xx.xxx:1234/v1",
"apiKey": "lmstudio",
"api": "openai-responses",
"models": [
{
"id": "openai/gpt-oss-120b",
"name": "GPT OSS 120B",
"reasoning": false,
"input": [
"text"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 128000,
"maxTokens": 8192
}
]
}
}
},

#

just replace the base url x's with the ip that lm studio server is running at. there's a local-models.md in the repo that also has a good walkthrough

wet spire
#

i don't get how you managed to get it to work, i even tried your config

#

and i get errors

#

Unhandled promise rejection: Error: Unhandled API in mapOptionsForApi: undefined

#

yea i went through the local models doc

#

i don't get how anyone was able to get it to work haha

#

i'm on latest version

#

so maybe they broke something

queen dew
#

try switching to "api": "openai-completions" maybe? and make sure something is in the API key field, doesn't matter what it is

wet spire
#

yea i tried both and made sure i have apiKey, do you mind sharing what version clawdbot you're on?

queen dew
#

2026.1.22

#

idk only other things that come to mind might be dumb but make sure the server is on and model is loaded. your config is the exact same?

wet spire
#

yes exact same and server is on and model is loaded

#

the only difference is i'm on a different version 2026.1.23-1

queen dew
#

can you send the full config?

wet spire
# queen dew can you send the full config? 
"models": {
    "mode": "merge",
    "providers": {
      "local": {
        "baseUrl": "http://127.0.0.1:1234/v1",
        "apiKey": "lmstudio",
        "api": "openai-responses",
        "models": [
          {
            "id": "openai/gpt-oss-20b",
            "name": "GPT OSS 20B",
            "reasoning": false,
            "input": [
              "text"
            ],
            "cost": {
              "input": 0,
              "output": 0,
              "cacheRead": 0,
              "cacheWrite": 0
            },
            "contextWindow": 128000,
            "maxTokens": 8192
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "local/openai/gpt-oss-20b"
      },
      "models": {
        "local/openai/gpt-oss-20b": {
          "alias": "gpt"
        },
        "anthropic/claude-opus-4-5": {
          "alias": "opus"
        },
        "anthropic/claude-3-5-haiku-latest": {
          "alias": "haiku"
        }
      },
      "workspace": "/Users/yui/clawd",
      "compaction": {
        "mode": "safeguard"
      },
      "maxConcurrent": 4,
      "subagents": {
        "maxConcurrent": 8
      }
    }
  },
#

i originally had the provider named lmstudio instead of local

#

but i updated it to local after seeing your config

#

but both fails

# 
05:51:57 [clawdbot] Unhandled promise rejection: Error: Unhandled API in mapOptionsForApi: undefined
    at mapOptionsForApi (file:///opt/homebrew/lib/node_modules/clawdbot/node_modules/@mariozechner/pi-ai/src/stream.ts:471:10)
    at streamSimple (file:///opt/homebrew/lib/node_modules/clawdbot/node_modules/@mariozechner/pi-ai/src/stream.ts:218:26)
    at streamAssistantResponse (file:///opt/homebrew/lib/node_modules/clawdbot/node_modules/@mariozechner/pi-agent-core/src/agent-loop.ts:233:25)
    at runLoop (file:///opt/homebrew/lib/node_modules/clawdbot/node_modules/@mariozechner/pi-agent-core/src/agent-loop.ts:141:20)
    at file:///opt/homebrew/lib/node_modules/clawdbot/node_modules/@mariozechner/pi-agent-core/src/agent-loop.ts:51:3
#

this is the error i get so it's not recognizing the api for some reason

#

i've tried openai-completions and openai-responses and openai-chat

queen dew
#

{
"models": {
"mode": "merge",
"providers": {
"local": {
"baseUrl": "http://127.0.0.1:1234/v1",
"apiKey": "lmstudio",
"api": "openai-responses",
"models": [
{
"id": "openai/gpt-oss-20b",
"name": "GPT OSS 20B",
"api": "openai-responses",
"reasoning": false,
"input": [
"text"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 128000,
"maxTokens": 8192
}
]
}
}
},
"agents": {
"defaults": {
"model": {
"primary": "local/openai/gpt-oss-20b"
},
"models": {
"local/openai/gpt-oss-20b": {
"alias": "gpt"
},
"anthropic/claude-opus-4-5": {
"alias": "opus"
},
"anthropic/claude-3-5-haiku-latest": {
"alias": "haiku"
}
},
"workspace": "/Users/yui/clawd",
"compaction": {
"mode": "safeguard"
},
"maxConcurrent": 4,
"subagents": {
"maxConcurrent": 8
}
}
}
}