WIth OpenAI FOrmat provider, succeded testing, but failed to chat

[L4zyD0g]

I want to use a third-party provider.
FIrst, I test the endpoint.

And curl also works.

But when I use chat, an error occured.

Did I miss something?

[L4zyD0g]

copilot log:

2026-01-31T08:04:44.023Z INFO [ChatManager] Sending message: "1234554321"
2026-01-31T08:04:44.024Z INFO [MessageRepository] Added message with ID: msg-1769846684016-kb8l8ovr4
2026-01-31T08:04:44.029Z INFO [UserMemoryManager] Saved Memories file not found, skipping saved memory load
2026-01-31T08:04:44.029Z INFO [ContextManager] Processing context for message msg-1769846684016-kb8l8ovr4
2026-01-31T08:04:44.031Z INFO [ContextManager] Successfully processed context for message msg-1769846684016-kb8l8ovr4
2026-01-31T08:04:44.031Z INFO [PromptContextEngine] Built envelope for message:msg-1769846684016-kb8l8ovr4 {"L1_SYSTEM":"e7686b54fa22ce276b9f8b0a1848f109e35fab0cc4d212d46fed871fad225051","L2_PREVIOUS":"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855","L3_TURN":"73b2f28465fdfe90170fab4445e3e1afa0c31debfeb270aa1fe4300448fdaca6","L4_STRIP":"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855","L5_USER":"eef1394ee3b80d9ce18ebfe618a056e6e0299c12cd19c99c796d482eb52277c6"}
2026-01-31T08:04:44.031Z INFO [MessageRepository] Updated processed text for message msg-1769846684016-kb8l8ovr4
2026-01-31T08:04:44.031Z INFO [ChatManager] Successfully sent message msg-1769846684016-kb8l8ovr4
2026-01-31T08:04:44.032Z INFO Step 0: Initial user message:\n 1234554321\n<active_note>\n<title>Untitled</title>\n<path>Untitled.md</path>\n<ctime>2026-01-31T08:01:13.552Z</ctime>\n<mtime>2026-01-31T08:01:13.552Z</mtime>\n<content>\n</content>\n</active_note>
2026-01-31T08:04:44.032Z INFO [LLMChainRunner] Using envelope-based context
2026-01-31T08:04:44.033Z INFO Final Request to AI:\n [{"role":"system","content":"<recent_conversations>\n ## Untitled Conversation\nTime: 2026-01-31 15:48\nSummary: Summary generation failed\n\n## Untitled Conversation\nTime: 2026-01-31 16:03\nSummary: Summary generation failed\n\n </recent_conversations>\n\n The current time is 2026-01-31 16:04.\n <recent_conversations> are the recent conversations between you and the user. \n You can use it to provide more context for your responses. \n Only use the recent conversations if they are relevant to the current conversation.\nYou are Obsidian Copilot, a helpful assistant that integrates AI to Obsidian note-taking.\n 1. Never mention that you do not have access to something. Always rely on the user provided context.\n 2. Always answer to the best of your knowledge. If you are unsure about something, say so and ask the user to provide more context.\n 3. If the user mentions "note", it most likely means an Obsidian note in the vault, not the generic meaning of a note.\n 4. If the user mentions "@vault", it means the user wants you to search the Obsidian vault for information relevant to the query. The search results will be provided to you in the context along with the user query, read it carefully and answer the question based on the information provided. If there's no relevant information in the vault, just say so.\n 5. If the user mentions any other tool with the @ symbol, check the context for their results. If nothing is found, just ignore the @ symbol in the query.\n 6. Always use $'s instead of \[ etc. for LaTeX equations.\n 7. When showing note titles, use [[title]] format and do not wrap them in .\n 8. When showing Obsidian internal image links, use ![[link]] format and do not wrap them in .\n 9. When showing web image links, use format and do not wrap them in .\n 10. When generating a table, format as github markdown tables, however, for table headings, immediately add ' |' after the table heading.\n 11. Always respond in the language of the user's query.\n 12. Do NOT mention the additional context provided such as getCurrentTime and getTimeRangeMs if it's irrelevant to the user message.\n 13. If the user mentions "tags", it most likely means tags in Obsidian note properties.\n 14. YouTube URLs: If the user provides YouTube URLs in their message, transcriptions will be automatically fetched and provided to you. You don't need to do anything special - just use the transcription content if available.\n 15. For markdown lists, always use '- ' (hyphen followed by exactly one space) for bullet points, with no leading spaces before the hyphen. Never use '*' (asterisk) for bullets."},{"role":"user","content":[{"type":"text","text":"<active_note>\n<title>Untitled</title>\n<path>Untitled.md</path>\n<ctime>2026-01-31T08:01:13.552Z</ctime>\n<mtime>2026-01-31T08:01:13.552Z</mtime>\n<content>\n\n</content>\n</active_note>\n\n---\n\n[User query]:\n\n1234554321"}]}]
2026-01-31T08:04:44.038Z INFO [ChatPersistenceManager] Created new chat file: copilot/copilot-conversations/1234554321@20260131_160444.md
2026-01-31T08:04:45.861Z ERROR [UserMemoryManager] Failed to extract title and summary: Connection error.\nmore message: Failed to fetch\nError: Connection error.\n at bn.makeRequest (plugin:copilot:402:3632)\n at async eval (plugin:copilot:410:24006)\n at async hQ (plugin:copilot:66:24134)\n at async i (plugin:copilot:65:7354)
2026-01-31T08:04:52.520Z ERROR [ChatPersistenceManager] Error generating AI topic: Connection error.\nmore message: Failed to fetch\nError: Connection error.\n at bn.makeRequest (plugin:copilot:402:3632)\n at async eval (plugin:copilot:410:24006)\n at async hQ (plugin:copilot:66:24134)\n at async i (plugin:copilot:65:7354)
2026-01-31T08:04:56.642Z ERROR Error during LLM invocation: Connection error.\nmore message: Failed to fetch
2026-01-31T08:04:56.642Z ERROR Connection error.\nmore message: Failed to fetch
2026-01-31T08:04:56.643Z INFO [MessageRepository] Added message with ID: msg-76aae6a5-b9ca-42f1-a398-17a7f4f3bb90
2026-01-31T08:04:56.643Z INFO Chat memory updated:\n {"turns":2}
2026-01-31T08:04:56.643Z INFO Final AI response (truncated):\n \n<errorChunk>Connection error.\nmore message: Failed to fetch</errorChunk>
2026-01-31T08:04:56.661Z INFO [ChatPersistenceManager] Updated existing chat file: copilot/copilot-conversations/1234554321@20260131_160444.md

Prompt — 2026-01-31T08:04:44.032Z

Actual Messages Sent to LLM:

[
  {
    "role": "system",
    "content": "<recent_conversations>\n        ## Untitled Conversation\n**Time:** 2026-01-31 15:48\n**Summary:** Summary generation failed\n\n## Untitled Conversation\n**Time:** 2026-01-31 16:03\n**Summary:** Summary generation failed\n\n        </recent_conversations>\n\n        The current time is 2026-01-31 16:04.\n        <recent_conversations> are the recent conversations between you and the user. \n        You can use it to provide more context for your responses. \n        Only use the recent conversations if they are relevant to the current conversation.\nYou are Obsidian Copilot, a helpful assistant that integrates AI to Obsidian note-taking.\n  1. Never mention that you do not have access to something. Always rely on the user provided context.\n  2. Always answer to the best of your knowledge. If you are unsure about something, say so and ask the user to provide more context.\n  3. If the user mentions \"note\", it most likely means an Obsidian note in the vault, not the generic meaning of a note.\n  4. If the user mentions \"@vault\", it means the user wants you to search the Obsidian vault for information relevant to the query. The search results will be provided to you in the context along with the user query, read it carefully and answer the question based on the information provided. If there's no relevant information in the vault, just say so.\n  5. If the user mentions any other tool with the @ symbol, check the context for their results. If nothing is found, just ignore the @ symbol in the query.\n  6. Always use $'s instead of \\[ etc. for LaTeX equations.\n  7. When showing note titles, use [[title]] format and do not wrap them in ` `.\n  8. When showing **Obsidian internal** image links, use ![[link]] format and do not wrap them in ` `.\n  9. When showing **web** image links, use ![link](url) format and do not wrap them in ` `.\n  10. When generating a table, format as github markdown tables, however, for table headings, immediately add ' |' after the table heading.\n  11. Always respond in the language of the user's query.\n  12. Do NOT mention the additional context provided such as getCurrentTime and getTimeRangeMs if it's irrelevant to the user message.\n  13. If the user mentions \"tags\", it most likely means tags in Obsidian note properties.\n  14. YouTube URLs: If the user provides YouTube URLs in their message, transcriptions will be automatically fetched and provided to you. You don't need to do anything special - just use the transcription content if available.\n  15. For markdown lists, always use '- ' (hyphen followed by exactly one space) for bullet points, with no leading spaces before the hyphen. Never use '*' (asterisk) for bullets."
  },
  {
    "role": "user",
    "content": [
      {
        "type": "text",
        "text": "<active_note>\n<title>Untitled</title>\n<path>Untitled.md</path>\n<ctime>2026-01-31T08:01:13.552Z</ctime>\n<mtime>2026-01-31T08:01:13.552Z</mtime>\n<content>\n\n</content>\n</active_note>\n\n---\n\n[User query]:\n\n1234554321"
      }
    ]
  }
]

Layered Context Metadata:

msg:msg-1769846684016-kb8l8ovr4 | conv:N/A | v1

━━━ SYSTEM MESSAGE ━━━

🔒 L1_SYSTEM (e7686b54) [CACHEABLE]
<recent_conversations>
        ## Untitled Conversation
**Time:** 2026-01-31 15:48
**Summary:** Summary generation failed

## Untitled Conversation
**Time:** 2026-01-31 16:03
**Summary:** Summary generation failed

        </recent_conversations>

        The current time is 2026-01-31 16:04.
      ...[truncated]

━━━ USER MESSAGE ━━━

⚡ L3_TURN (73b2f284)
<active_note>
<title>Untitled</title>
<path>Untitled.md</path>
<ctime>2026-01-31T08:01:13.552Z</ctime>
<mtime>2026-01-31T08:01:13.552Z</mtime>
<content>

</content>
</active_note>

⚡ L5_USER (eef1394e)
1234554321

Settings

{
  "userId": "6167729b-51d9-4006-98a6-05ae555642cd",
  "isPlusUser": false,
  "amazonBedrockRegion": "",
  "githubCopilotAccessToken": "",
  "githubCopilotToken": "",
  "githubCopilotTokenExpiresAt": 0,
  "defaultChainType": "llm_chain",
  "defaultModelKey": "copilot-plus-flash|copilot-plus",
  "embeddingModelKey": "text-embedding-3-small|openai",
  "temperature": 0.1,
  "maxTokens": 6000,
  "contextTurns": 15,
  "userSystemPrompt": "",
  "openAIProxyBaseUrl": "",
  "openAIEmbeddingProxyBaseUrl": "",
  "stream": true,
  "defaultSaveFolder": "copilot/copilot-conversations",
  "defaultConversationTag": "copilot-conversation",
  "autosaveChat": true,
  "generateAIChatTitleOnSave": true,
  "autoAddActiveContentToContext": true,
  "defaultOpenArea": "view",
  "defaultSendShortcut": "enter",
  "customPromptsFolder": "copilot/copilot-custom-prompts",
  "indexVaultToVectorStore": "ON MODE SWITCH",
  "qaExclusions": "copilot",
  "qaInclusions": "",
  "chatNoteContextPath": "",
  "chatNoteContextTags": [],
  "enableIndexSync": true,
  "debug": true,
  "enableEncryption": false,
  "maxSourceChunks": 15,
  "enableInlineCitations": true,
  "activeModels": [
    {
      "name": "copilot-plus-flash",
      "provider": "copilot-plus",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "plusExclusive": true,
      "projectEnabled": false,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "google/gemini-2.5-flash-lite",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "google/gemini-2.5-flash",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "google/gemini-2.5-pro",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gpt-5",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gpt-5-mini",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gpt-4.1",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gpt-4.1-mini",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "core": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "openai/gpt-4.1",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": false,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "openai/gpt-4.1-mini",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": false,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "grok-4-fast",
      "provider": "xai",
      "enabled": true,
      "isBuiltIn": true,
      "core": false,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "x-ai/grok-4-fast",
      "provider": "openrouterai",
      "enabled": true,
      "isBuiltIn": true,
      "core": false,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "claude-sonnet-4-20250514",
      "provider": "anthropic",
      "enabled": true,
      "isBuiltIn": true,
      "capabilities": [
        "reasoning",
        "vision"
      ]
    },
    {
      "name": "gemini-2.5-flash",
      "provider": "google",
      "enabled": true,
      "isBuiltIn": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gemini-2.5-flash-lite",
      "provider": "google",
      "enabled": true,
      "isBuiltIn": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "gemini-2.5-pro",
      "provider": "google",
      "enabled": true,
      "isBuiltIn": true,
      "projectEnabled": true,
      "capabilities": [
        "vision"
      ]
    },
    {
      "name": "azure-openai",
      "provider": "azure openai",
      "enabled": true,
      "isBuiltIn": true
    },
    {
      "name": "deepseek-chat",
      "provider": "deepseek",
      "enabled": true,
      "isBuiltIn": true
    },
    {
      "name": "deepseek-reasoner",
      "provider": "deepseek",
      "enabled": true,
      "isBuiltIn": true,
      "capabilities": [
        "reasoning"
      ]
    },
    {
      "name": "deepseek-ai/DeepSeek-V3",
      "provider": "siliconflow",
      "enabled": false,
      "isBuiltIn": true,
      "baseUrl": "https://api.siliconflow.com/v1"
    },
    {
      "name": "deepseek-ai/DeepSeek-R1",
      "provider": "siliconflow",
      "enabled": false,
      "isBuiltIn": true,
      "baseUrl": "https://api.siliconflow.com/v1",
      "capabilities": [
        "reasoning"
      ]
    },
    {
      "name": "ark-code-latest",
      "provider": "3rd party (openai-format)",
      "enabled": true,
      "isBuiltIn": false,
      "baseUrl": "https://ark.cn-beijing.volces.com/api/coding/v3",
      "isEmbeddingModel": false,
      "capabilities": [
        "reasoning",
        "websearch"
      ],
      "stream": true
    }
  ],
  "activeEmbeddingModels": [
    {
      "name": "copilot-plus-small",
      "provider": "copilot-plus",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true,
      "plusExclusive": true
    },
    {
      "name": "copilot-plus-large",
      "provider": "copilot-plus-jina",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true,
      "plusExclusive": true,
      "believerExclusive": true,
      "dimensions": 1024
    },
    {
      "name": "copilot-plus-multilingual",
      "provider": "copilot-plus-jina",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true,
      "plusExclusive": true,
      "dimensions": 512
    },
    {
      "name": "text-embedding-3-small",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true
    },
    {
      "name": "gemini-embedding-001",
      "provider": "google",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true
    },
    {
      "name": "Qwen/Qwen3-Embedding-0.6B",
      "provider": "siliconflow",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true,
      "core": true,
      "baseUrl": "https://api.siliconflow.com/v1"
    },
    {
      "name": "text-embedding-3-large",
      "provider": "openai",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true
    },
    {
      "name": "embed-multilingual-light-v3.0",
      "provider": "cohereai",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true
    },
    {
      "name": "text-embedding-004",
      "provider": "google",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true
    },
    {
      "name": "azure-openai",
      "provider": "azure openai",
      "enabled": true,
      "isBuiltIn": true,
      "isEmbeddingModel": true
    }
  ],
  "embeddingRequestsPerMin": 60,
  "embeddingBatchSize": 16,
  "disableIndexOnMobile": true,
  "showSuggestedPrompts": true,
  "showRelevantNotes": true,
  "numPartitions": 1,
  "lexicalSearchRamLimit": 100,
  "promptUsageTimestamps": {},
  "promptSortStrategy": "timestamp",
  "chatHistorySortStrategy": "recent",
  "projectListSortStrategy": "recent",
  "defaultConversationNoteName": "{$topic}@{$date}_{$time}",
  "inlineEditCommands": [],
  "projectList": [],
  "lastDismissedVersion": null,
  "passMarkdownImages": true,
  "enableAutonomousAgent": true,
  "enableCustomPromptTemplating": true,
  "enableSemanticSearchV3": false,
  "enableLexicalBoosts": true,
  "suggestedDefaultCommands": true,
  "autonomousAgentMaxIterations": 4,
  "autonomousAgentEnabledToolIds": [
    "localSearch",
    "readNote",
    "webSearch",
    "pomodoro",
    "youtubeTranscription",
    "writeToFile",
    "replaceInFile",
    "updateMemory"
  ],
  "reasoningEffort": "low",
  "verbosity": "medium",
  "memoryFolderName": "copilot/memory",
  "enableRecentConversations": true,
  "maxRecentConversations": 30,
  "enableSavedMemory": true,
  "quickCommandIncludeNoteContext": true,
  "autoIncludeTextSelection": false,
  "autoAddSelectionToContext": false,
  "autoAcceptEdits": false,
  "diffViewMode": "split",
  "userSystemPromptsFolder": "copilot/system-prompts",
  "defaultSystemPromptTitle": "",
  "autoCompactThreshold": 128000
}

[logancyang]

cc @Emt-lin could you pls take a look when you get a chance? thanks!

[Emt-lin]

cc @Emt-lin could you pls take a look when you get a chance? thanks!

yeah