Skip to main content

hearth

Deterministic model residency. Keep declared models warm, and tell the truth about which ones are.

Not an inference engine. llama.cpp and vLLM have spent years on kernels, samplers and tokenizers, and none of that is the problem. The problem is that no serving stack will promise a model stays loaded, and none of them can say why one stopped being.

The night this came from

A PIN operator on one rented RTX A6000 stopped answering. Requests hung, then failed. Four pull requests landed against the streaming path in a single evening and none of them were the cause — because the cause was never visible. Three completely different failures were arriving as the same timeout:

  • the runtime evicted a model to free VRAM,
  • the host detached the GPU and gave it to another tenant,
  • a 32B model was simply still loading.

One of those is a capacity problem you own. One is your provider's and no configuration will touch it. One is not a problem at all. A timeout cannot tell you which, so all three got "fixed" repeatedly and none of them went away.

Worse, the operator was being scored down for a card their host reclaimed. A reputation system fed that kind of data slowly deletes its own honest operators.

What it does

$ hearth status
42.0 / 44.2 GiB held  (5 declared, 2 admitted)
  muse-local:latest            resident for 14203s
  deepseek-r1:32b              loading for 47s
  gemma4:26b                   not admitted — short by 14.8 GiB
  qwen3.6:27b                  not admitted — short by 15.8 GiB
  gemma4-extract:31b           not admitted — short by 17.8 GiB

Three things, none of which you can get today:

1. Residency is a named state with a named reason.

Unknown    never probed
Loading    weights materializing — with elapsed, so it can say how long
Resident   loaded AND answering AND accounted for
Lost       with a reason: Evicted · GpuDetached · ProcessExited · Unhealthy
Failed     won't load, and what the runtime said
Stopped    unloaded on purpose

Evicted and GpuDetached are the two states nothing else reports, and they were the two that mattered. They are distinguished by one bit — whether the GPU was still present when the probe failed — and that bit is the entire diagnosis.

2. The card's size is arithmetic, checked before anything loads.

Declare four 20 GiB models on a 48 GiB card and no runtime errors. It loads, evicts, loads, evicts, forever, and presents as "the models got slow." hearth refuses the fifth model at declare time and tells you it was short by 17.8 GiB. Nothing is ever evicted to make room for a load — if it doesn't fit, the honest answer is that it doesn't fit.

3. Routers get an answer they can act on.

answer what a router should do
Ready send it
Warming{for_ms} wait, or try elsewhere — but do not fault this node
Lost{GpuDetached} try elsewhere, and do not score this operator down
Lost{Evicted} try elsewhere; this box is over-committed
NotAdmitted{short} stop asking — it will never fit here
Unknown we genuinely don't know yet, and we say so

Design rules

Nothing fails on a clock. A 32B materializing over a network fabric can legitimately spend minutes before its first token, and killing it at an arbitrary deadline turns a slow success into a fast failure. Loading reports how long it has been loading; deciding what to do about that belongs to whoever knows if a human is waiting. Progress is reported — patience is a policy, not a constant.

Loading is not Ready. The most common way a serving stack lies is routing to something still coming up and calling the inevitable timeout an error.

Declaration order is priority order. First fit, never best fit. Reordering to squeeze in one more model would silently demote whatever the operator listed first, and on a serving box first means most important. A planner that outsmarts the operator surprises them at 3am.

The reserve is never planned into. Weights aren't the whole cost — KV cache grows with context and parallelism, the CUDA context is hundreds of megabytes, and fragmentation is real on a card that's been up for weeks.

Status

hearth-core — state machine, VRAM planner, fleet routing. Pure logic, no GPU required. hearth-resolve — one reference syntax over the Ollama registry and HuggingFace GGUFs, verified against the live registries. hearth-storethe NEDB spine. Every residency transition is a bi-temporal, causally-linked, tamper-evident event in an embedded NEDB. Not a log on the side: the supervisor's memory IS the database. hearth-serve — the supervisor. llama-server children, honest health probes, gpu_present via nvidia-smi (an honest None on CPU boxes), SIGTERM records unloaded and reaps children. Plus the hearth CLI.

Proven end-to-end on a real model: hearth serve brought a GGUF resident under a real llama-server (warmup measured, not guessed), served real completion tokens, took a SIGTERM, and a fresh process then read the whole story back off disk:

$ hearth why stories
● seq     3  unloaded       {}
└─ seq     2  resident       {"endpoint":"127.0.0.1:18080","warmup_ms":256}
└─ seq     1  loading        {"endpoint":"127.0.0.1:18080","pid":10121}
└─ seq     0  declared       {"admitted":true,"vram_bytes":1073741824}

$ hearth as-of stories 2
as of seq 2: stories was `resident` {"endpoint":"127.0.0.1:18080","warmup_ms":256}

$ hearth verify
verify ok — 4 nodes checked, history intact

What was resident as of seq 2 is a real query against a real causal chain. When a model goes cold at 3am you get the answer instead of a theory — and verify proves nobody rewrote it. No other serving stack can print that.

Next: hearth pull wired to the resolvers + blob store · HTTP surface (OpenAI-compatible proxy + /residency) · multi-model fleets under one supervisor · napi + PyO3 bindings for store/serve.

Integration

hearth speaks OpenAI-compatible, so pin-clientd works with it today: set apiMode: "openai" and point inferenceUri at hearth. /residency then adds the truth that the OpenAI shape has no way to express.

Build

cargo test --workspace   # no GPU needed
cargo run --example a6000

# serve a model under supervision (needs llama.cpp's llama-server)
hearth serve --model muse --gguf ./muse.gguf --port 8080
hearth status && hearth why muse && hearth verify

© Interchained LLC · BUSL-1.1 (converts to Apache-2.0 on 2030-08-27)

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hearth_engine-0.3.1.tar.gz (35.4 kB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

hearth_engine-0.3.1-cp39-abi3-win_amd64.whl (205.3 kB view details)

Uploaded CPython 3.9+Windows x86-64

hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (349.9 kB view details)

Uploaded CPython 3.9+manylinux: glibc 2.17+ x86-64

hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (338.2 kB view details)

Uploaded CPython 3.9+manylinux: glibc 2.17+ ARM64

hearth_engine-0.3.1-cp39-abi3-macosx_11_0_arm64.whl (300.3 kB view details)

Uploaded CPython 3.9+macOS 11.0+ ARM64

hearth_engine-0.3.1-cp39-abi3-macosx_10_12_x86_64.whl (308.3 kB view details)

Uploaded CPython 3.9+macOS 10.12+ x86-64

File details

Details for the file hearth_engine-0.3.1.tar.gz.

File metadata

  • Download URL: hearth_engine-0.3.1.tar.gz
  • Upload date:
  • Size: 35.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: maturin/1.15.0

File hashes

Hashes for hearth_engine-0.3.1.tar.gz
Algorithm Hash digest
SHA256 ac5f726019d92a18b87472188c7d5dfa7a52b5f5476244e817c6130d8b2195b8
MD5 5121c2bf8abb2e70666928c3d366eeb4
BLAKE2b-256 2e8019531a4f78cb4d174eac295d8ce0e7c2488dabeacf4494d1fce2af84e3e8

See more details on using hashes here.

File details

Details for the file hearth_engine-0.3.1-cp39-abi3-win_amd64.whl.

File metadata

File hashes

Hashes for hearth_engine-0.3.1-cp39-abi3-win_amd64.whl
Algorithm Hash digest
SHA256 9e3d04950419bdddde8e0c1b0e8ad57e24fe26e19dd421f0b79ad2f33dec92c2
MD5 1ad7a0cecfd8caac19f82c6db92dd302
BLAKE2b-256 a1d1de53ac4469c16b825a3c24fa024e8e96398f6e413c9a6ce03a9e85390897

See more details on using hashes here.

File details

Details for the file hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 f582c20de7b1e079618738409e048ec55173dd4ab6a8b28cdefbe9ef58ca3114
MD5 f2e92132debc0265aba86e0579b34890
BLAKE2b-256 63e65a04784a72ed7eb85eb440d0678b9706b76073acdf8b9ee8f958ffe92d66

See more details on using hashes here.

File details

Details for the file hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.

File metadata

File hashes

Hashes for hearth_engine-0.3.1-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Algorithm Hash digest
SHA256 2ecd8782fe0ca1234f09221632cddad8d8c284e3c701dd16ccb56696d399b1fd
MD5 918107b3be31cc1c4d83f62085f9cc94
BLAKE2b-256 76e7acb6fc6321f9f763728277fcd49e994bd862fe32fb52087fe98d5f0a5be1

See more details on using hashes here.

File details

Details for the file hearth_engine-0.3.1-cp39-abi3-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for hearth_engine-0.3.1-cp39-abi3-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 8d1749f0811e6c1c01d7cbd6873144deef614cda6c50931748faef8954025005
MD5 116aee354ea761bde2eeb4f6a23c5387
BLAKE2b-256 ab910ef6a9268137cf41e6349b02b79d6d993a2392e0aa0bcf4bc385e39831db

See more details on using hashes here.

File details

Details for the file hearth_engine-0.3.1-cp39-abi3-macosx_10_12_x86_64.whl.

File metadata

File hashes

Hashes for hearth_engine-0.3.1-cp39-abi3-macosx_10_12_x86_64.whl
Algorithm Hash digest
SHA256 bd621222666a8413038a314c8ff89dd033d3d4cbba2195d63c3a68de6160fa6d
MD5 87a2d9f7663f7ec6b5eb7070a17df2a2
BLAKE2b-256 a6e8893d9bcb01b739691010b6881bb5bfbb60705a7f927c220203cd0f561710

See more details on using hashes here.

Release history Release notifications | RSS feed

0.4.7

6 files

0.4.6

6 files

0.4.5

6 files

0.4.3

6 files

0.4.2

6 files

0.4.1

6 files

0.4.0

6 files

0.3.9

6 files

0.3.8

6 files

0.3.7

6 files

0.3.6

6 files

0.3.5

6 files

0.3.4

6 files

0.3.3

6 files

0.3.2

6 files

This release

0.3.1 This release

6 files

0.3.0

6 files

0.2.0

6 files

0.1.2

6 files

0.1.1

6 files

0.1.0

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page