Llm Cost Python Packages

helicone-helpers

🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

inferrail

Self-hosted OpenAI/Anthropic-compatible gateway: give each AI agent run a dollar budget, enforced before the provider, with a payload-free per-run cost record.

ferrogw

Unified AI Gateway for 30+ LLMs (OpenAI, Anthropic, Bedrock, Azure etc) with Caching, Guardrails, A/B test & cost controls. Go-native Fastest & Scalable AI Gateway LiteLLM & Kong AI Gateway alternative.

tokenmeter-dashboard

Claude Code token usage and cost dashboard. Cost per prompt, cache misses and usage limits for Claude Code, Codex and GitHub Copilot CLI, read from local logs. Zero dependencies, nothing leaves your machine.

adk-tracegauge

See what your Google ADK agent costs (per-invocation USD from token usage) and fail CI when it gets more expensive.

replay-doctor

If the agent bill went up and nothing errored, a prompt cache broke. Replay Doctor names the turn it broke on, the cause, and the tokens re-billed at write prices. Reads Claude Code and Codex transcripts already on disk. Local Go CLI, zero third-party dependencies, no account, and no network request you did not type.

cloud-finops-mcp

Cloud FinOps Skill & MCP by OptimNow. FinOps Foundation aligned knowledge for AI agents: cloud cost (AWS/Azure/GCP/OCI), AI inference economics, Kubernetes, data platforms, allocation, chargeback, anomaly management, waste detection, GreenOps. Installs into 12 AI tools, or connect the hosted MCP server by URL. Refreshed twice a month.

kalibr

Stop overpaying to run your agents. Kalibr routes every request to lower-cost model and tool paths without degrading performance.

tokencur

The CUR for your tokens — convert multi-provider AI/LLM usage into FOCUS-conformant cost datasets, validated by the FinOps Foundation's focus-validator

helicone

🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

claude-router

claude-router is a local prompt router that picks the right Claude model tier and prepends the right scaffold using local embeddings before you call the API. A deterministic routing layer for eval, research, content, and review prompts that helps teams stop overspending on Sonnet and Opus when Haiku plus structure is enough.

kry-attest

Tamper-evident, stranger-verifiable receipts for LLM cost-savings — anyone can recompute your caching/routing savings math offline, no trust in your dashboard required. Pure stdlib, zero-dependency.

prompt-receipts

What an AI coding run cost, took and did: receipts for your own Claude Code, Codex and OpenCode sessions, plus prompts run on 7 models

ai-netcafe

Zero-dependency toolbox for AI agents: web search, page-to-markdown, diagrams, transcription, PDF, cross-session memory, and one-sentence app deployment. No install, no signup to start.

tracegauge

Three-axis efficiency scorer for Claude Code sessions.

token-police

TokenPolice Python SDK

tokentrust-cli

CLI and MCP server that verifies AI coding agent context-reduction proxies' token and cost savings claims

baar-core

Budget-Aware Agentic Routing (BAAR) — Intelligent LLM model selection with a zero-call financial kill-switch. Save 90% on costs without losing accuracy.

argrouter

LLM router: picks the model and reasoning effort per query from accuracy and real billed cost.

tellan

Reconcile LLM API token usage and costs with local estimates, tamper-evident ledgers, and audit statements.

claude-code-cost-explorer

Local-first observability for Claude Code - drill into costs, prompts, and tool calls turn by turn. Zero instrumentation.

mcp-budget-governor

Distributed, cost-denominated budget enforcement for MCP servers: per-user quotas, per-tool limits, and a global spend circuit breaker backed by atomic Redis counters.

helicone-async

🧊 Open source LLM observability platform. One line of code to monitor, evaluate, and experiment. YC W23 🍓

llmkit-sdk

Know what your AI agents cost. API gateway with budget enforcement, session tracking, and MCP tools.