[OpenAI] Responses API support for Node.js

[codefromthecrypt]

Is your feature request related to a problem? Please describe.

The OpenAI platform released an alternative to the completions endpoint called responses last month

https://platform.openai.com/docs/api-reference/responses

They are now using this by default in the playground (which generates content) as well as the new codex api written in typescript https://github.com/openai/codex

I've successfully run this with EDOT SDK Node.js, however I only get HTTP spans for the operations because it, like the OpenAI Agents SDK, uses the responses api instead of the completions one.

diff --git a/codex-cli/package.json b/codex-cli/package.json index 38850e6..67d9707 100644 --- a/codex-cli/package.json +++ b/codex-cli/package.json @@ -10,6 +10,7 @@ "node": ">=22" }, "scripts": { + "codex": "node build.mjs --dev && node --env-file .env -r @elastic/opentelemetry-node ./dist/cli-dev.js", "format": "prettier --check src tests", "format:fix": "prettier --write src tests", "dev": "tsc --watch", @@ -43,7 +44,8 @@ "openai": "^4.89.0", "react": "^18.2.0", "shell-quote": "^1.8.2", - "use-interval": "1.4.0" + "use-interval": "1.4.0", + "@elastic/opentelemetry-node": "^1" }, "devDependencies": { "@eslint/js": "^9.22.0",

Describe the solution you'd like

Logs, metrics and tracing for responses like we do for completions.

TODO feature checkboxes

Describe alternatives you've considered

Try to modify https://github.com/openai/codex to support completions instead

Additional context

announcement: https://community.openai.com/t/introducing-the-responses-api/1140929

local inference support of core features

• litellm https://github.com/BerriAI/litellm/releases/tag/v1.67.0-stable
• llama-stack https://github.com/meta-llama/llama-stack/releases/tag/v0.2.4
• masaic-ai https://github.com/masaic-ai-platform/open-responses/releases/tag/v0.1.3-M2
• ollama Support for OpenAI Responses API (for Codex CLI compatibility) ollama/ollama#10309
• vllm [Feature]: Support openai responses API interface vllm-project/vllm#14721

other javascript/typescript/node SDKs

• openinference https://www.npmjs.com/package/@arizeai/openinference-instrumentation-openai/v/2.3.0

This is the same issue as EDOT Python SDK elastic/elastic-otel-python-instrumentations#62

[codefromthecrypt]

FYI I updated this as now there is one released node instrumentation library that claims to support this (openinference), as well litellm claims to support the server-side.

Note: A project owner will need to further edit this issue description to break down what support means. For example, if it means the same features as needed by openai codex, or something else. Also, if there are inaccuracies or anything that becomes out of date in the issue description, a lead likely want to edit in corrections.

[codefromthecrypt]

note openinference supports this in their node.js SDK and 3 different local inference platforms (though not ollama, yet) support it as well.

[codefromthecrypt]

note: I've verified that at least at a basic level, llama-stack and open-responses emulate the responses API when proxying to ollama.

There may be other instrumentation and platform updates besides this, but I wanted to give this one as I can personally verify they work.