Get an audit

Head to head · Local inference · October 2026 research run

KoboldCpp vs Lemonade

Lemonade scores 63.8 (B) on agent readiness against KoboldCpp's 60.5 (C), and leads in 4 of 7 scored categories. KoboldCpp leads on maintenance & community. Both do local inference.

KoboldCppLostRuins (Concedo) · HTTP API C60.5/100#462 of 841 LemonadeAMD and the Lemonade community · HTTP API B63.8/100#336 of 841

Which one, for what

KoboldCpp C

Good for An owner who wants text, image, speech and music models behind one executable with a writing and roleplay interface, and clients that speak the KoboldAI, OpenAI, Ollama or Anthropic formats.

Ahead on

  • Maintenance & community, 82 against 76

Also in its favour

  • No key needed to call it

Watch for

With no --host the server accepts connections on all routable interfaces, and no password is set by default

Lemonade B

Good for An owner with AMD hardware (Ryzen AI NPUs, Radeon or Strix Halo) who wants one local server for chat, speech and images that existing OpenAI, Anthropic or Ollama clients can call, and for MCP clients that want local models as tools.

Ahead on

  • Reliability, 75 against 68
  • Agent ergonomics, 70 against 63
  • Transparency & trust, 64 against 49

Watch for

No authentication by default. With no key set, every route answers, including /internal/* shutdown and configuration

Score by category

CategoryWeight this runKoboldCppLemonadeEdge
Reliability16%206875Lemonade +7
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.26870Lemonade +2
Agent ergonomics13%16.26370Lemonade +7
Security & auth14%17.53836KoboldCpp +2
Payments & pricing10%12.56060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.88276KoboldCpp +6
Transparency & trust7%8.84964Lemonade +15
Negative events≤1500
Total60.5 · C63.8 · B

Facts side by side

FactKoboldCppLemonade
KindHTTP APIHTTP API
VendorLostRuins (Concedo)AMD and the Lemonade community
Hosted endpointno (local only)no (local only)
TransportsHTTPHTTP
AuthNoneAPI key
PricingFreeFree
x402nono
LicenceAGPL-3.0 for KoboldCpp and KoboldAI Lite. The bundled GGML, llama.cpp and stable-diffusion.cpp code stays under MITApache 2.0. Each backend (llama.cpp, whisper.cpp, stable-diffusion.cpp, FastFlowLM and others) is downloaded separately under its own licence
Tools exposednone6
Read-only variant documentednono
llms.txtyesno
Last release2026-09-272026-10-07
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Customer content may train models
Terms restrict automated access
Terms restrict benchmarking
Terms or service can change without notice
Arbitration or class-action waiver
Popularity12k stars5.8k stars

Verdicts

KoboldCpp

One file runs text, image, speech and music models behind a published OpenAPI 3.0.3 document, with eight releases in 90 days. The server listens on every interface with no password by default, and --password leaves the image routes open.

Lemonade

Apache 2.0, with weekly releases, installers for Windows, macOS and five Linux routes, and a six-tool MCP endpoint whose descriptions steer a caller away from accidental multi-gigabyte downloads. Authentication is off by default, the repository has no security policy, and the docs tell WebSocket clients to pass the key in the URL.

Before you call either

KoboldCpp

  1. Start with --host 127.0.0.1 and --password. The default listens on every interface with no key
  2. Send the password as Authorization: Bearer <password>. It is not read from the query string
  3. Treat 503 as both busy and rate limited. The server never sends 429 or Retry-After, and the wait in seconds is in detail.msg
  4. Pass max_length or max_tokens. The default is 2,048 tokens unless --defaultgenamt changes it
  5. Send a genkey with each generation so /api/extra/generate/check and /api/extra/abort act on your request and not another caller's

Lemonade

  1. Call lemonade_list_models (or GET /v1/models) before naming a model. A wrong name with allow_download: true can start a multi-gigabyte download
  2. Send Authorization: Bearer <key> when the operator has set LEMONADE_API_KEY. /internal/* needs the admin key when one is set
  3. Use POST /v1/chat/completions for streamed tokens, embeddings and speech. The MCP endpoint ignores stream and has no embeddings or text-to-speech tool
  4. Pass output_dir to lemonade_generate_image and lemonade_omni to get file paths in place of inline base64. Writes stay inside the MCP media sandbox
  5. Read GET /v1/docs on the running server for the reference that matches its version. Weekly releases change behaviour

Questions

Which is better for AI agents, KoboldCpp or Lemonade?

Lemonade scores 63.8 (B) on agent readiness against KoboldCpp's 60.5 (C), and leads in 4 of 7 scored categories. KoboldCpp leads on maintenance & community.

Do KoboldCpp and Lemonade need an API key?

KoboldCpp needs no key. Lemonade needs an API key.

Can an agent call KoboldCpp and Lemonade without installing anything?

No hosted endpoint is listed for KoboldCpp. No hosted endpoint is listed for Lemonade.

Are KoboldCpp and Lemonade open source?

Yes. KoboldCpp is open source (AGPL-3.0 for KoboldCpp and KoboldAI Lite. The bundled GGML, llama.cpp and stable-diffusion.cpp code stays under MIT). Lemonade is open source (Apache 2.0. Each backend (llama.cpp, whisper.cpp, stable-diffusion.cpp, FastFlowLM and others) is downloaded separately under its own licence).

Other comparisons with KoboldCpp or Lemonade

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an auditSee a sample report