Write a review

Head to head · Local inference · October 2026 research run

Berth vs MLX LM

Berth scores 67.6 (B) on agent readiness against MLX LM's 52.5 (D), and leads in 5 of 7 scored categories. Both do local inference.

Best local AI models and assistants · All 229 local ai comparisons

BerthTrevin Tindall (T-Crypt on GitHub) · HTTP API B67.6/100#283 of 1218 MLX LMApple Inc. · HTTP API D52.5/100#866 of 1218

Which one, for what

Berth B

Best for Someone running several GPU programs (llama.cpp, llama-swap, Ollama, vLLM, ComfyUI, games) on one NVIDIA card who wants them to take turns, and agents that need to ask for the card.

Ahead on

  • Reliability, 76 against 66
  • Schema & documentation, 69 against 37
  • Agent ergonomics, 75 against 54
  • Security & auth, 51 against 32
  • Maintenance & community, 76 against 61

Watch for

First released on 11 October 2026 and called alpha. Windows, AMD and multiple GPUs are experimental, and the gateway is a preview

MLX LM D

Best for An owner with an Apple silicon Mac who wants MLX-format models, local fine-tuning and quantisation from Python or the command line, with a simple local chat completions server.

Also in its favour

  • No key needed to call it

Watch for

mlx_lm.server has no API key or other credential option, and --allowed-origins defaults to *

Score by category

CategoryWeight this runBerthMLX LMEdge
Reliability16%207666Berth +10
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.26937Berth +32
Agent ergonomics13%16.27554Berth +21
Security & auth14%17.55132Berth +19
Payments & pricing10%12.56060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87661Berth +15
Transparency & trust7%8.86869MLX LM +1
Negative events≤1500
TotalB 67.6/100D 52.5/100

Facts side by side

FactBerthMLX LM
KindHTTP APIHTTP API
VendorTrevin Tindall (T-Crypt on GitHub)Apple Inc.
Hosted endpointno (local only)no (local only)
TransportsHTTPHTTP
AuthAPI keyNone
PricingFreeFree
x402nono
LicenceApache-2.0. The Berth name, the Pilot mascot and the artwork in assets/ are excluded from the licence, per the NOTICE fileMIT
Read-only variant documentednono
llms.txtnono
Last release2026-10-232026-10-01
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
Popularity245 stars7.3k stars, 140k PyPI/wk

Verdicts

Berth

Berth coordinates one NVIDIA card between engines, ComfyUI and games, grants a lease only after the driver shows memory free, and gives agents a skill, JSON output and a keyed loopback gateway. It is 17 days old, alpha and maintained by one developer, with Windows and AMD experimental and no machine-readable API spec.

MLX LM

MIT, with no telemetry found in the source, and the tests passed on the last eight pushes to main. mlx_lm.server has no API key option, answers any origin by default and loads whichever model a request names, and its own docs say it is not recommended for production.

Before you call either

Berth

  1. Run berth status --format json before starting, stopping or switching any GPU engine. If berthd is unreachable, no engine is protected
  2. Never kill a process to free VRAM. Ask for the card with berth run, berth hold or a gateway request, and report stowaways to the user
  3. Send gateway requests to http://127.0.0.1:9799/v1 with Authorization: Bearer $(berth key), and use "stream": true for cold starts of up to minutes
  4. On a 503, read the code and wait for Retry-After. A pinned holder (a game or berth hold) answers 409 to stop and is the user's call
  5. Ask the user before berth reclaim LEASE --yes, any berth.toml edit or berth update. Each changes the node

MLX LM

  1. Keep mlx_lm.server on 127.0.0.1 and pass --allowed-origins with the origins you trust. There is no API key, and the default answers every origin
  2. Treat any caller as able to load any model. The model and adapters request fields accept any Hugging Face repository or local path
  3. Send max_tokens or max_completion_tokens when you need more than 512 tokens, the server default
  4. Read errors as {"error": "<text>"} with 400 for a bad field and 404 for a model that failed to load. They are not OpenAI error objects
  5. Poll GET /health before the first request. It answers 503 with unavailable when the generation thread has stopped

Questions

Which is better for AI agents, Berth or MLX LM?

Berth scores 67.6 (B) on agent readiness against MLX LM's 52.5 (D), and leads in 5 of 7 scored categories.

Do Berth and MLX LM need an API key?

Berth needs an API key. MLX LM needs no key.

Can an agent call Berth and MLX LM without installing anything?

No hosted endpoint is listed for Berth. No hosted endpoint is listed for MLX LM.

Are Berth and MLX LM open source?

Yes. Berth is open source (Apache-2.0. The Berth name, the Pilot mascot and the artwork in assets/ are excluded from the licence, per the NOTICE file). MLX LM is open source (MIT).

Other comparisons with Berth or MLX LM

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