LiteLLM alternatives (2026)

LiteLLM is a agent infra tool: Open-source AI gateway and Python SDK that calls 100+ LLM providers in the OpenAI format, with cost tracking, guardrails, load balancing and logging. Pricing, as the maker states it: Open source + cloud. Its GitHub repo has 61,465 stars. We found 33 tools in the directory that do overlapping work (building AI agents, building voice agents, building MCP servers); 24 of them are open source and 3 list a free or freemium plan. Below are the 10 closest, ranked by how many tasks and sub-tasks they share with LiteLLM, then by GitHub stars.

Updated October 21, 2026

Quick comparison Top 10 of 11 listed below. Pricing as the maker states it; blank means we don't have it.

Tool	Pricing	Best for	Open source
NVIDIA Open Agent Safety Platform	Open source	Monitor & evaluate agents	Yes
opencodex	Open source	Write code	Yes
Portkey AI Gateway	Freemium
Manifest	Open source + cloud	Automate workflows	Yes
Open Connector	Open source	Build an agent	Yes
Monid	Open source + cloud	Voice & phone calls	Yes
Kong Gateway	Open source + cloud	Connect tools via MCP	Yes
Arcade	Open source + cloud	Build an agent	Yes
Agent Barn	Open source + cloud	Monitor & evaluate agents	Yes
TypeSafe Jev	Usage-based: pay per input token, output free; Jev 1.14 listed at $0.042 per million input tokens in TypeSafe docs; also on OpenRouter; enterprise plans via sales	Write code

1. NVIDIA Open Agent Safety Platform (Open source): Reference stack pairing OpenShell with NVIDIA Sentry on BlueField-4 DPUs for hardware-level agent monitoring. Consider it because it also covers terminal and CLI agents.
2. opencodex (Open source): A local provider proxy that lets OpenAI Codex (CLI, app and SDK) and Claude Code run on other model providers such as Claude, Gemini, Grok, DeepSeek or local Ollama models. Consider it because it also covers local and self-hosted and terminal and CLI agents.
3. Portkey AI Gateway (Freemium): Fast open-source AI gateway that routes requests to hundreds of LLMs through one API with fallbacks, load balancing and built-in guardrails. Consider it because it does the same job with a different approach (freemium).
4. Manifest (Open source + cloud): Open-source LLM gateway that routes each request from agents and apps to the right model across API keys, subscriptions and local models, with spend tracking. Consider it because it also covers local and self-hosted.
5. Open Connector (Open source): Open-source auth gateway connecting SaaS apps to AI agents via SDK, CLI and MCP. Consider it because it also covers identity, auth and secrets and terminal and CLI agents.
6. Monid (Open source + cloud): Gateway that gives an agent one base URL and one key to call thousands of paid tools, from web scraping and enrichment to image and voice generation, billed per call. Consider it because it also covers scraping and crawling and images and design.
7. Kong Gateway (Open source + cloud): Cloud-native API, LLM and MCP gateway that adds routing, auth, multi-LLM support, semantic security and MCP traffic analytics through plugins. Consider it because it also covers identity, auth and secrets.
8. Arcade (Open source + cloud): Actions runtime for enterprise agents with thousands of prebuilt authenticated tools, plus an open-source framework for building your own MCP servers. Consider it because it also covers identity, auth and secrets and MCP servers.
9. Agent Barn (Open source + cloud): Agent Barn is an open-source (Apache-2.0) control plane for running a fleet of AI agents on your own Kubernetes: create, configure, run and monitor agents across Slack, Microsoft Teams and Telegram from one place. Consider it because it also covers running many agents in parallel and agents in chat apps.
10. TypeSafe Jev (Usage-based: pay per input token, output free; Jev 1.14 listed at $0.042 per million input tokens in TypeSafe docs; also on OpenRouter; enterprise plans via sales): Jev is TypeSafe's System One decision model, built for the small decisions inside software and agents. You send a state and typed questions (choice, score and more) and get typed answers with probabilities back instead of free-form text, so code can branch on them directly to route, classify, score or rank options. Agents can use it to settle narrow calls cheaply and save the larger LLM for open-ended work. Available through TypeSafe's API, Python and JavaScript SDKs, an agent skill, and OpenRouter. Consider it because it does the same job with a different approach (usage-based: pay per input token, output free; jev 1.14 listed at $0.042 per million input tokens in typesafe docs; also on openrouter; enterprise plans via sales).

LiteLLM vs alternatives at a glance

Tool	Pricing	Open source / licence	GitHub stars	Good for
LiteLLM	Open source + cloud	MIT	61,465	building AI agents, building voice agents
NVIDIA Open Agent Safety Platform	Open source	Yes	—	coding in the terminal, evaluating agents
opencodex	Open source	Yes	17,682	writing code, coding in the terminal
Portkey AI Gateway	Freemium	MIT	13,214	agent security and safety
Manifest	Open source + cloud	MIT	7,595	automating workflows, evaluating agents
Open Connector	Open source	Apache License 2.0	6,031	coding in the terminal, automating workflows
Monid	Open source + cloud	Yes	5,298	scraping websites, building voice agents
Kong Gateway	Open source + cloud	Apache License 2.0	44,278	connecting tools with MCP, building MCP servers
Arcade	Open source + cloud	Yes	1,054	building AI agents, running business operations
Agent Barn	Open source + cloud	Yes	—	evaluating agents, monitoring agents in production
TypeSafe Jev	Usage-based: pay per input token, output free; Jev 1.14 listed at $0.042 per million input tokens in TypeSafe docs; also on OpenRouter; enterprise plans via sales	No	—	writing code, building AI agents

More: LiteLLM listing · All agent infra tools · All alternatives pages

Alternatives are picked from directory data only: same category plus shared tasks and sub-tasks. Pricing is as each maker states it; GitHub stars as recorded on each listing.