What the person above is suggesting: * https://pi.dev/ * https://omp.sh/ Personally I also think that OpenCode is nice, their CLI version is enjoyable and their desktop/web version is mostly okay: * https://opencode.ai/. - Source: Hacker News / 3 days ago

I wired it into OpenCode, an open-source coding harness similar to Claude Code and Codex. I'm using an OpenCode Zen key and a Gemini key, and together, these give me access to multiple SOTA models for free, without touching a separate dashboard for each provider. - Source: dev.to / 13 days ago

I drove it from my coding-agent with a handful of commands. - Source: dev.to / 16 days ago

Https://pi.dev/docs/latest/providers#openai-codex is specific that it "Requires ChatGPT Plus or Pro subscription". https://opencode.ai are also both specific that it's for "Plus/Pro" subscriptions. Is there some link where you saw free tier Codex use in non-official apps is allowed by OpenAI? (I have used the free tier in the official Codex app, but as you said, labs can have different rules for official vs... - Source: Hacker News / 19 days ago

Https://opencode.ai/ OpenCode was the first agent harness I used, and I have always like it. You can configure a wide variety of providers, but it's open source and has a number of core contributors. The other opinionated option is Pi (the Pi agent harness). This is a great lightweight option and also supports a number of providers. You can also use local model servers. - Source: Hacker News / about 2 months ago

Langfuse and LangSmith exist for this. Use them. The 30 minutes you spend setting up observability saves you the 87 hours you'd spend debugging blind. - Source: dev.to / 8 days ago

In this project we will build a Python banking assistant agent using Strands Agents and make it observable and continuously evaluated using Langfuse — step by step. - Source: dev.to / about 2 months ago

Langfuse is the open-source standard for LLM observability. It traces every LLM interaction — prompts, completions, latency, token usage, cost — and provides the tooling to debug, evaluate, and optimize LLM applications in production. Think of it as "Datadog for LLM calls" with a focus on prompt engineering workflows. - Source: dev.to / 2 months ago

You're monitoring production traffic. You need Langfuse / Phoenix / Helicone / Braintrust for that. Online eval is a different problem class: implicit feedback, drift detection, hallucination rates on your data, not on HellaSwag. - Source: dev.to / 3 months ago

Gateway or proxy attribution. A reverse proxy in front of the model-provider API records the request, computes the cost, and exposes per-customer breakdowns. Open-source options include Helicone, LiteLLM, Langfuse, and OpenLLMetry. Hosted equivalents serve as the AI cost observability layer for teams that want centralized visibility: LangSmith, Datadog LLM Observability, Arize Phoenix. Adds a network hop.... - Source: dev.to / 3 months ago