No output in Opencode with OpenAI-compatible Straico provider (tokens used, cURL works, UI empty)

When using the Straico OpenAI-compatible API as a provider in Opencode (v0.11.1), prompts are received and tokens are billed on the Straico backend, but no assistant response is ever surfaced in the Opencode UI. There is no error, no output, and no “message” content displayed. This occurs with all tested Straico chat models, including x-ai/grok-4, openai/o3-mini, anthropic/claude-sonnet-4, etc.

cURL produces correct, complete responses.
Session, provider, and model are detected and initialized in logs.
Opencode logs show request sent, model found, “getSDK” completed, “start-step”/“finish-step” published—but no content displayed.

To Reproduce

Set up opencode.json with confirmed model ID(s) from Straico /v0/models endpoint.
Authenticate opencode auth login (straico chosen).
Start Opencode with opencode --print-logs
Send a prompt via UI/chat.
Dialogue never appears, session completes, no errors visible.

Expected behavior
Assistant reply (message.content) from Straico API should be shown in UI.

Screenshots / Logs

Straico /v0/models returns correct model IDs.

cURL to Straico:

curl -X POST "https://api.straico.com/v0/chat/completions" \
-H "Authorization: Bearer <MY_API_KEY>" \
-H "Content-Type: application/json" \
-d '{"model": "x-ai/grok-4", "messages": [{"role":"user", "content":"Test message"}]}'

Gives response like:

{
  "choices": [{
    "message": { "role": "assistant", "content": "Hello! This is Grok, built by xAI, responding..." }
  }],
  ...
}

Opencode logs (snippet):

service=provider providerID=straico modelID=x-ai/grok-4 found
... type=start-step part
... type=finish-step part
... type=finish part

No error, no output, session times out/ends normally.

Additional context

Removal of cache (rm -rf ~/.cache/opencode ~/.local/share/opencode) and latest release make no difference.
Occurs across all tested Straico OpenAI-compatible models.
Model usage and billing confirmed in Straico dashboard.
cURL always returns working output (see above).

System details

Opencode version: 0.11.1
macOS
Straico API, OpenAI-compatible REST
Models tested: x-ai/grok-4, openai/o3-mini, anthropic/claude-sonnet-4 and more

can u share your opencode.json?

Sure. I have added this json file under /Users/nites/.config/opencode on my mac.

{
"$schema": "https://opencode.ai/config.json",
"provider": {
"straico": {
"npm": "@ai-sdk/openai-compatible",
"options": {
"baseURL": "https://api.straico.com/v0"
},
"models": {
"amazon/nova-lite-v1": {},
"amazon/nova-micro-v1": {},
"amazon/nova-pro-v1": {},
"anthropic/claude-3-opus": {},
"anthropic/claude-3-5-haiku-20241022": {},
"anthropic/claude-3.5-sonnet": {},
"anthropic/claude-3.7-sonnet:thinking": {},
"anthropic/claude-3.7-sonnet": {},
"anthropic/claude-sonnet-4": {},
"cohere/command-r-08-2024": {},
"cohere/command-r-plus-08-2024": {},
"deepseek/deepseek-chat": {},
"deepseek/deepseek-r1": {},
"deepseek/deepseek-r1:nitro": {},
"deepseek/deepseek-chat-v3-0324": {},
"deepseek/deepseek-chat-v3.1": {},
"cognitivecomputations/dolphin-mixtral-8x7b": {},
"alpindale/goliath-120b": {},
"google/gemini-2.5-flash": {},
"google/gemini-2.5-flash-lite": {},
"google/gemini-2.0-flash-001": {},
"google/gemini-pro-1.5": {},
"google/gemini-2.5-pro-preview": {},
"google/gemma-2-27b-it": {},
"gryphe/mythomax-l2-13b": {},
"meta-llama/llama-3-70b-instruct:nitro": {},
"meta-llama/llama-3.1-405b-instruct": {},
"meta-llama/llama-3.1-70b-instruct": {},
"meta-llama/llama-3.3-70b-instruct": {},
"meta-llama/llama-4-maverick": {},
"microsoft/phi-4": {},
"mistralai/codestral-mamba": {},
"mistralai/mistral-medium-3": {},
"mistralai/mixtral-8x7b-instruct": {},
"moonshotai/kimi-k2:free": {},
"moonshotai/kimi-k2-0905": {},
"nvidia/llama-3.1-nemotron-70b-instruct": {},
"nvidia/llama-3.1-nemotron-ultra-253b-v1": {},
"nvidia/llama-3.3-nemotron-super-49b-v1": {},
"openai/gpt-4.1": {},
"openai/gpt-4.1-mini": {},
"openai/gpt-4.1-nano": {},
"openai/gpt-4o-2024-08-06": {},
"openai/gpt-4o-2024-11-20": {},
"openai/gpt-4o-mini": {},
"openai/gpt-5": {},
"openai/gpt-5-chat": {},
"openai/gpt-5-mini": {},
"openai/gpt-5-nano": {},
"openai/o1": {},
"openai/o1-pro": {},
"openai/o1-mini": {},
"o3-2025-04-16": {},
"o3-deep-research-2025-06-26": {},
"openai/o3-mini-high": {},
"openai/o3-mini": {},
"openai/o4-mini": {},
"openai/o4-mini-high": {},
"perplexity/llama-3.1-sonar-small-128k-online": {},
"perplexity/sonar": {},
"perplexity/sonar-deep-research": {},
"perplexity/sonar-reasoning": {},
"qwen/qwen-2-72b-instruct": {},
"qwen/qwen-2-vl-72b-instruct": {},
"qwen/qwen-2.5-72b-instruct": {},
"qwen/qwen2.5-vl-32b-instruct:free": {},
"qwen/qwen3-235b-a22b": {},
"qwen/qwen3-coder": {},
"microsoft/wizardlm-2-8x22b": {},
"z-ai/glm-4.5v": {},
"x-ai/grok-2-1212": {},
"x-ai/grok-3-beta": {},
"x-ai/grok-3-mini-beta": {},
"x-ai/grok-4": {}
}
}
}
}

Hmm I think i will need to create a straico account to test this I tried using some other providers (setup using same mechanism u are doing) and it worked fine for me

I will get back to you once I have setup straico or encountered similar issues to u

Thank you for looking into this. If you are on Discord Opencode, please share your username and I will share test API of Straico with you, if that's helpful

aidenc9727

Just shared via Discord

I tested the following for straico:

import { createOpenAICompatible } from "@ai-sdk/openai-compatible";
import { streamText } from "ai";

const straico = createOpenAICompatible({
  name: "straico",
  baseURL: "https://api.straico.com/v0",
  apiKey: "****",
});

const model = straico("openai/gpt-4.1-nano");

const result = streamText({
  model,
  prompt: "hello",
});

for await (const textPart of result.textStream) {
  console.log(textPart);
}

This returns nothing so streaming may not be properly implemented on their end?

They don't seem to properly follow openai spec so this isn't an opencode issue seems like a bug on their end

Hi, it’s Nicolás from Straico Support. I just wanted to share that we recently added Chat Completion and Model Information endpoints to our API v2, which are fully OpenAI-compatible. Streaming isn’t supported yet, but asynchronous completions work normally with OpenAI libraries, so previous code should work by replacing the streaming function with the chat completion function.

For anyone interested, check out our documentation here: https://documenter.getpostman.com/view/5900072/2s9YyzddrR#f16ebe47-10a4-41eb-b2e2-97831e7ee6de

Navigation Menu

Repository navigation

No output in Opencode with OpenAI-compatible Straico provider (tokens used, cURL works, UI empty) #2724

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

No output in Opencode with OpenAI-compatible Straico provider (tokens used, cURL works, UI empty) #2724

Description

Activity

rekram1-node commented on Sep 21, 2025

niteshmanav commented on Sep 21, 2025

rekram1-node commented on Sep 21, 2025

niteshmanav commented on Sep 22, 2025

rekram1-node commented on Sep 22, 2025

niteshmanav commented on Sep 22, 2025

rekram1-node commented on Sep 23, 2025

nicolas23589 commented on Oct 18, 2025

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions