Get an audit

Head to head · Local inference · October 2026 research run

Docker Model Runner vs KoboldCpp

KoboldCpp scores 60.5 (C) on agent readiness against Docker Model Runner's 57.1 (C), and leads in 3 of 7 scored categories. Docker Model Runner leads on reliability and transparency & trust. Both do local inference.

Docker Model RunnerDocker, Inc. · HTTP API C57.1/100#559 of 841 KoboldCppLostRuins (Concedo) · HTTP API C60.5/100#462 of 841

Which one, for what

Docker Model Runner C

Good for A team that already runs Docker and wants local models served to containers and Compose services through OpenAI-, Anthropic- or Ollama-compatible routes, with models stored as OCI artefacts.

Ahead on

  • Reliability, 85 against 68
  • Transparency & trust, 73 against 49

Watch for

No credential on the API. The docs say any client that can reach it, including other containers, can pull, load and run models

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

  • Schema & documentation, 68 against 49
  • Agent ergonomics, 63 against 58
  • Maintenance & community, 82 against 55

Also in its favour

  • No incidents deducted, where Docker Model Runner loses 3 points for them

Watch for

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

Score by category

CategoryWeight this runDocker Model RunnerKoboldCppEdge
Reliability16%208568Docker Model Runner +17
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.24968KoboldCpp +19
Agent ergonomics13%16.25863KoboldCpp +5
Security & auth14%17.54038Docker Model Runner +2
Payments & pricing10%12.56060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.85582KoboldCpp +27
Transparency & trust7%8.87349Docker Model Runner +24
Negative events≤15-30
Total57.1 · C60.5 · C

Facts side by side

FactDocker Model RunnerKoboldCpp
KindHTTP APIHTTP API
VendorDocker, Inc.LostRuins (Concedo)
Hosted endpointno (local only)no (local only)
TransportsHTTPHTTP
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache-2.0 (server, CLI plugin and dmr binary). Docker Desktop, which bundles it, is closed software under Docker's subscription agreement, and each model carries its own licenceAGPL-3.0 for KoboldCpp and KoboldAI Lite. The bundled GGML, llama.cpp and stable-diffusion.cpp code stays under MIT
Read-only variant documentednono
llms.txtyesyes
Last release2026-08-122026-09-27
Terms last updated2026-08-26no document linked
Privacy policy last updated2026-08-26no document linked
Customer content may train modelsnot found in the text
Terms restrict automated accessyes
Terms restrict benchmarkingyes
Terms or service can change without noticenot found in the text
Arbitration or class-action waiveryes
Popularity656 stars12k stars

Verdicts

Docker Model Runner

CI passes on the main branch, and Docker has published two security advisories with CVEs and fixed versions for the project. The API takes no credential, so any client or container that reaches it can pull, delete and run models, and the documentation has no OpenAPI file or error reference.

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.

Before you call either

Docker Model Runner

  1. Use base URL http://localhost:12434/engines/v1 for OpenAI clients and http://localhost:12434 for Anthropic and Ollama clients. Any API key value is accepted
  2. In Docker Desktop, run docker desktop enable model-runner --tcp 12434 first. Host-side TCP is off by default
  3. From a container, call http://model-runner.docker.internal on Docker Desktop or http://172.17.0.1:12434 on Docker Engine
  4. Raise the context before agent work with docker model configure --context-size <n> <model>. The llama.cpp default is 4,096 tokens
  5. Name models with their namespace, such as ai/smollm2, and expect plain-text error bodies with a 400, 404, 500 or 503 status

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

Questions

Which is better for AI agents, Docker Model Runner or KoboldCpp?

KoboldCpp scores 60.5 (C) on agent readiness against Docker Model Runner's 57.1 (C), and leads in 3 of 7 scored categories. Docker Model Runner leads on reliability and transparency & trust.

Do Docker Model Runner and KoboldCpp need an API key?

Neither needs a key.

Can an agent call Docker Model Runner and KoboldCpp without installing anything?

No hosted endpoint is listed for Docker Model Runner. No hosted endpoint is listed for KoboldCpp.

Are Docker Model Runner and KoboldCpp open source?

Yes. Docker Model Runner is open source (Apache-2.0 (server, CLI plugin and `dmr` binary). Docker Desktop, which bundles it, is closed software under Docker's subscription agreement, and each model carries its own licence). 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).

Other comparisons with Docker Model Runner or KoboldCpp

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