Self-hosted LLM operations tools

Figure and bar: visits/mo, estimate · year: online since

1. #14G

   GreenPT

   Provides AI chat, APIs, and specialized tools running on European renewable energy infrastructure with transparent energy tracking.

   28k
   Active
   From ≈ $5.06/mo
   2025
2. #19L

   LiteLLM

   Routes requests across 140+ LLM providers and 1800+ models through a single API gateway with cost optimization and access controls.

   15.2k
   Active
   Free plan
   2023
3. #22L

   Liquid AI

   Distributes small foundation models optimized to run on devices without sending data to servers.

   11.3k
   Active
   Free plan
   2023
4. #30B

   Baseten

   Deploys and serves AI models in production with optimized inference across cloud infrastructure.

   5.9k
   Active
   Free plan
   2019
5. #42O

   OmniRoute

   Routes requests across 352 model providers through a single OpenAI-compatible endpoint, falling back automatically when a provider is unavailable or out of quota.

   641
   Active
6. #53E

   EvoX Genesis

   GPU-accelerated runtime for self-evolving agents, models, and workflows with evolutionary algorithms.

   132
   Active
7. #58I

   Inferencer

   Runs and inspects large language models locally with token-level control and distributed inference.

   77
   Active
   From $9.99/mo
8. #80N

   NoemaAI Launch

   Runs language models locally on Apple devices with file search and tool integration, syncing across Mac, iPhone, iPad and Vision Pro.

   20
   Active
   Free
9. #88L

   Lekh AI

   Runs large language models, generates images and videos, and creates music entirely on-device without internet or data collection.

   16
10. #89O

    Outlier

    Runs large language models locally on Apple Silicon Macs for coding, file editing, and research without sending data to servers.

    16
    Active
    From $249 once
11. #91L

    LiberaGPT

    Runs large language models entirely offline on iPhone and Android devices without subscriptions or data collection.

    14
    Active
    Free
12. #106G

    Gridlight

    Runs AI inference on your own hardware with model routing, cost governance, and data residency controls.

    3
    Active
    Contact sales
13. #108B

    Bodhi App

    Unified gateway for running local language models, cloud APIs, and MCP tools through a single chat interface.

    3
    Active
14. #113B

    Bike4MindMCP · open

    Runs chat, agent jobs and interactive artifacts across hosted frontier models and self-hosted open-weight models from one workbench.

    3
    Active
    From $30/mo
15. #119M

    mlx-serve

    Runs open-source language models and generative tools locally on Mac, with chat, image generation, voice cloning, and agent capabilities.

    2
    Active
16. #130B

    Base Compute

    Optimizes large language models to run locally on device hardware with minimal latency and cost.

    1
    Active
17. #131F

    Fortaify

    Runs language models, image generation, and voice processing locally on your device without sending data to the cloud.

    1
    Active
    From $8.99/mo
18. #148P

    PandaProbe

    Traces agent runs, evaluates outcomes, and generates validated fixes so agents can learn from failures autonomously.

    99 linking domains
    Active
    From $29/mo
19. #150Y

    Your Own AI

    Runs large language models locally on your machine with optional cloud routing for frontier models.

    60 linking domains
    Active
    From $200/yr
20. #153C

    co/core

    Runs open-source AI models on members' shared hardware using a standard API compatible with OpenAI clients.

    44 linking domains
    Active
21. #158V

    varCHAR

    Trains domain-specific language models inside your network without exporting data.

    39 linking domains
    Active
    From $15,000 once
    2025
22. #165A

    AI Cost Gate

    Self-hosted gateway that logs metadata of LLM calls and enforces per-project and per-agent spend budgets.

    27 linking domains
    Active
    From $29 once
23. #171Z

    zizka.ai

    Stores agent execution traces, replays sessions, and detects behavior drift in production.

    29 linking domains
    Active
24. #179K

    KernelAI

    Runs 53 open-source language models locally on your phone with no account or internet required.

    20 linking domains
    Active
25. #185D

    Dryade

    Orchestrates language model agents on your servers with policy-driven workflows, tool routing, and cryptographic audit trails.

    Emerging
    Active
    From ≈ $55/user/mo
26. #188D

    Decentralized AI Protocol/Launchpad

    Decentralized AI protocol where miners earn rewards for GPU compute and developers launch AI models on Solana.

    Emerging
27. #195J

    jitllm

    Runs LLM inference by compiling kernels for your CPUs and GPUs and paging model blocks from disk as needed.

    52 ★
    Active
28. #198R

    Ryuksaidso MVP

    Manages agent execution with tracing, approvals, and multi-provider failover for production AI systems.

    Little sign of use
    Active
29. L

    Libra

    Runs compressed large language models locally on Apple silicon Macs using the MLX framework.

    1
    Active
    From ≈ $12/mo
30. T

    Titen

    Stores agent memory as sourced observations and claims, compiles budgeted context packs, and serves them over an API and MCP.

    1.6k installs/mo
    Active