Langfuse Alternative for Coding-Agent Cost per Task
Langfuse is a full LLM observability platform. If what you need is the cost of each finished Claude Code or Codex task, AgentReceipt does that with no SDK and no traces of your prompts.
What Langfuse is
Langfuse is an open-source LLM engineering platform: tracing, evals, prompt management and cost tracking for applications you instrument with its SDK.
Where Langfuse is the better fit:
- Deep request-level tracing for your own LLM apps
- Evals and prompt management
- Self-hostable
What AgentReceipt does differently
- AgentReceipt needs no SDK: it reads the logs coding agents already write
- Prompts and code never leave the machine; Langfuse stores full traces
- Built around the finished task (PR, ticket), with cost explanations and savings
- Priced per workspace at $19/mo rather than by events
Pricing
Langfuse Cloud has a free Hobby tier, then Core from $29/mo and Pro from $199/mo plus usage (per its pricing page, 2026).
Product details change; check each vendor’s site for current features and prices. Many teams use both: an observability platform for their own LLM apps, and AgentReceipt for coding-agent spend per task.
Itemize your own session
Drop a Claude Code or Codex log. It is parsed in your browser, and you get your real numbers instead of a model.
Drop a session log here
Claude Code ~/.claude/projects/…/*.jsonl, a Codex rollout ~/.codex/sessions/…/rollout-*.jsonl, or the claude-code-action execution file.
Parsed in your browser. Your prompts and code never leave this tab.