@ai-sdk/azure provider does not work with /chat/completions since v1.3.4

Description

@ai-sdk/azure provider does not work since v1.3.4

here is my opencode.json config

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "foundary": {
      "npm": "@ai-sdk/azure",
      "options": {
        "baseURL": "https://custom-domain/openai",
        "apiKey": "xxxxxxxx",
        "apiVersion": "2025-04-01-preview",
        "useDeploymentBasedUrls": true,
        "useCompletionUrls": true
      },
      "models": {
        "gpt-5.4": {
          "name": "GPT-5.4"
        }
      }
    }
  },
  "autoupdate": false
}

I have looked up a lot of similar issue regarding azure openai api, seems like everyone is confuse about azure multiple supporting API format.

Expect:
POST: https://custom-domain/openai/deployments/{deployment-id}/chat/completions?api-version={api-version}

{
  "model": "gpt-5.4",
  "messages": [
    ...
  ],
  ...
}

Actual:
POST: https://custom-domain/openai/deployments/{deployment-id}/responses?api-version={api-version}

{
  "model": "gpt-5.4",
  "input": [
    ...
  ],
  ...
}

This line of code sdk.chat(modelID) should direct me to the chat completion method with useCompletionUrls options based on the logic

changing @ai-sdk/azure version to 3.0.49 in opencode v1.3.4 might cause this issue to appear

Plugins

No response

OpenCode version

v1.3.4

Steps to reproduce

prepare a azure api that have (only) this API format available - https://custom-domain/openai/deployments/{deployment-id}/chat/completions?api-version={api-version}
use opencode version v1.3.3 with this config - it works
use opencode version v1.3.4 with same config - it does not works

Screenshot and/or share link

No response

Operating System

macOS

Terminal

WezTerm

are u getting a 404 not found? I managed to hit azure completions just fine

are u getting a 404 not found? I managed to hit azure completions just fine

I did get Resource Not Found error in response.

I try running debug log and saw it was hitting /response endpoint, which indicates it was using response method from @ai-sdk/azure provider.

I also try forcing the api path with similar to this method, but later found out that the request body json is in different formats with the different error message I received.

One thing that I am uncertain is that in previous opencode versions, I did not even input useCompletionUrls option at all and the api still works fine, hitting the completion endpoint correctly. But based on the code logic, it should default to using response instead.

Oh I think I found it

Wait why is ur provider called "foundary"

If it was called "azure" I think it'd work but it's not?

"useCompletionUrls" is something we have it's only tied to that oen provider name

fix: respect useCompletionUrls for custom @ai-sdk/azure providers

Wait why is ur provider called "foundary"

If it was called "azure" I think it'd work but it's not?

It does work after I change the namespace from foundary to azure (tested on opencode v1.3.13)

  "provider": {
    "azure": {
      "npm": "@ai-sdk/azure",
      "options": {
        "baseURL": "https://custom-domain/openai",
        "apiKey": "xxxxxxxx",
        "apiVersion": "2025-04-01-preview",
        "useDeploymentBasedUrls": true,
        "useCompletionUrls": true
      },
      "models": {
        "gpt-5.4": {
          "name": "GPT-5.4"
        }
      }
    }
  }

I thought this is just a display name tag in TUI model select, separating custom provider with default provider, didn't know it have actual mapping functionality

Does this mean provider[key] and provider[key].npm always need to pair up correctly?

No, it's just that deployment / completion urls thing is kinda an old artifact, im not even sure if it is documented there has got to be a cleaner solution.

But it has been hard coded specifically for that provider this whole time

We have azure support but it's not like great in terms of UX and accounting for all the different setups so if u have any ideas we are all ears

@chunglam2525

@chunglam2525 @rekram1-node I am sorry but is this broken again? I can't get it to work in 1.4.11 with azure as provider name and useCompletionUrls set to true.

@peri4n

@peri4n are you experiencing this error?

Function tools with reasoning_effort are not supported for gpt-5.5 in /v1/chat/completions. Please use /v1/responses instead.

In opencode v1.3.16, this pr fixes provider options for azure sdk. But it breaks the /v1/chat/completions method.

Removing reasoning_effort from fetch body does solve this problem, but I am not sure this will be a good solution, since this is too much of custom logic for this specific provider. @rekram1-node

packages/opencode/src/provider/provider.ts

// Strip openai itemId metadata following what codex does
if (
  (model.api.npm === "@ai-sdk/openai" || model.api.npm === "@ai-sdk/azure") &&
  opts.body &&
  opts.method === "POST"
) {
  const body = JSON.parse(opts.body as string)
  const keepIds = body.store === true
  let bodyChanged = false
  if (!keepIds && Array.isArray(body.input)) {
    for (const item of body.input) {
      if ("id" in item) {
        delete item.id
        bodyChanged = true
      }
    }
  }
  // Remove reasoning_effort for /v1/chat/completions
  if (options["useCompletionUrls"]) {
    delete body["reasoning_effort"]
    bodyChanged = true
  }
  if (bodyChanged) {
    opts.body = JSON.stringify(body)
  }
}

For now, I would recommend sticking to a fix version of opencode if you are using azure with /chat/completion endpoint, since it is already mention in above reply that this azure deployment/completion endpoint being an old artifact with discontinue updates

@chunglam2525

Hi @chunglam2525,
thank you for the response.

I get a 404 Ressource not Found because I can't opencode to actually use a chat/completions endpoint even with useCompletionUrls set to true.

Hi @chunglam2525, thank you for the response.

I get a 404 Ressource not Found because I can't opencode to actually use a chat/completions endpoint even with useCompletionUrls set to true.

Can you share your provider config? it seems like it is reaching an incorrect endpoint.
Or what you could do is using DEBUG log opencode --log-level DEBUG to investigate the API call from opencode.

maybe share opencode debug config

I hope this helps:

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "git": {
      "command": [
        "/nix/store/n6704zw3zhjmzf4xh37lvq40p914rzk6-mcp-server-git-2026.1.26/bin/mcp-server-git"
      ],
      "enabled": true,
      "type": "local"
    },
    "nixos": {
      "command": [
        "/nix/store/29zsf22sip45hm3nkmkw0myp40kc27hb-mcp-nixos-2.4.3/bin/mcp-nixos"
      ],
      "enabled": true,
      "type": "local"
    },
    "sequential-thinking": {
      "command": [
        "/nix/store/6b0rp1chaakhv2si1z7mik890n0w50aw-mcp-server-sequential-thinking-2026.1.26/bin/mcp-server-sequential-thinking"
      ],
      "enabled": true,
      "type": "local"
    }
  },
  "model": "enbw_anthropic/Latest",
  "provider": {
    "azure": {
      "models": {
        "gpt-5": {
          "name": "GPT-5"
        },
        "gpt-5-mini": {
          "name": "GPT-5 Mini"
        }
      },
      "name": "ENBW OpenAI",
      "npm": "@ai-sdk/azure",
      "options": {
        "apiKey": "API_KEY",
        "apiVersion": "2025-02-01-preview",
        "baseURL": "https://example.com/openai",
        "useCompletionUrls": true,
        "useDeploymentBasedUrls": true
      }
    },
    "enbw_anthropic": {
      "models": {
        "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
          "name": "Latest"
        }
      },
      "name": "ENBW Anthropic",
      "npm": "@ai-sdk/amazon-bedrock",
      "options": {
        "apiKey": "API_KEY",
        "baseURL": "https://example.com/anthropic",
        "headers": {
          "api-key": "API_KEY"
        },
        "region": "eu-central-1"
      }
    }
  },
  "agent": {},
  "mode": {},
  "plugin": [
    "file:///home/fbull/.config/opencode/plugin/env-protection.js"
  ],
  "command": {},
  "plugin_origins": [
    {
      "spec": "file:///home/fbull/.config/opencode/plugin/env-protection.js",
      "source": "/home/fbull/.config/opencode",
      "scope": "global"
    }
  ],
  "username": "fbull"
}

Navigation Menu

Repository navigation

@ai-sdk/azure provider does not work with /chat/completions since v1.3.4 #20287

Description

Plugins

OpenCode version

Steps to reproduce

Screenshot and/or share link

Operating System

Terminal

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

@ai-sdk/azure provider does not work with /chat/completions since v1.3.4 #20287

Description

Description

Plugins

OpenCode version

Steps to reproduce

Screenshot and/or share link

Operating System

Terminal

Activity

rekram1-node commented on Mar 31, 2026

chunglam2525 commented on Mar 31, 2026

rekram1-node commented on Mar 31, 2026

rekram1-node commented on Mar 31, 2026

rekram1-node commented on Mar 31, 2026

chunglam2525 commented on Apr 1, 2026

rekram1-node commented on Apr 1, 2026

rekram1-node commented on Apr 1, 2026

peri4n commented on Apr 23, 2026

chunglam2525 commented on May 6, 2026

peri4n commented on May 7, 2026

chunglam2525 commented on May 11, 2026

rekram1-node commented on May 11, 2026

peri4n commented on May 11, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions