A model is still running

A model is still running or downloading. Leaving will stop the current operation. Stay on this page to let it finish.

Skip to content
LocalAI.toolsBeta

Join the waitlist

Get an email when the new apps and model runs launch.

No account needed.

Liquid LFM3 Desktop

Local text chat with Liquid LFM3 1.7B.

Desktop inferencePrivate · Runs on your computer
About this model

Continue a literal text or code prefix locally with the pretrained Liquid LFM3 1.7B Base checkpoint.

Created by
Liquid AI
Model released
Unknown
Source repository created
Oct 26, 2026
Model weights · FP32
3.5 GB

Best for

  • Short prose continuation
  • Short code continuation

Keep in mind

  • This adapter continues literal prefixes. It does not accept chat messages, tools or media.
  • Ambiguous prefixes can produce inaccurate or unsafe continuations. Verify facts and review code before execution.
  • The adapter caps input and output at 8192 tokens combined. Long-context, multilingual and domain-specific quality are unverified.
  • The model license limits commercial use based on annual revenue.

Processed on your deviceYour input is processed locally in the desktop app. Model and runtime files download from external hosts.

Target platforms · verification required

  • macos · aarch64, x86_64
  • windows · x86_64

Recommended configuration

FP32 · mps (verified platform only)

Sources & technical details
Original model
https://huggingface.co/LiquidAI/LFM3-1.7B-Base
Release verification
Unknown
Release evidence
The official model card identifies this exact checkpoint. The retained repository creation time does not establish its public release date.
Source updated
Nov 3, 2026
Native model files
LiquidAI/LFM3-1.7B-Base
Files updated
Nov 3, 2026
Artifact revision
7bca5038823c0d0f14efe06439109354a10e014c

Desktop app

Sample result

LFM3 1.7B Base Desktop

Run Liquid LFM3 Desktop on your computer

Download the desktop app to load this model, provide input and view results locally.

Public desktop downloads are coming soon.

Device requirements

macOS Apple SiliconCPU / FP32Tested14.9 GiBRAM estimate
macOS Apple SiliconMPS / FP32Tested14.9 GiBRAM estimate
macOS Apple SiliconCPU / BF16Tested14.9 GiBRAM estimate
macOS Apple SiliconMPS / BF16Tested14.9 GiBRAM estimate
macOS Apple SiliconMPS / FP16Tested14.9 GiBRAM estimate
If your device falls short

MemoryAdvisory only: close other apps, reduce image/tile/token/batch settings or choose a smaller verified model/precision. CPU can need more RAM than GPU.

GPUIf the selected GPU runtime fails, existing CPU fallback may recover when enabled, but CPU has its own memory/runtime needs.

PlatformUse a compatible OS/architecture/runtime or another model. These estimates do not add hardware detection or block runs.

Benchmark evidence

Saved desktop measurements apply to the recorded hardware and runtime. Fallback cases report the device that actually ran; they do not prove GPU support on the requested device.

LFM3 1.7B Base Desktop

5 configurations tested
Passed5
Failed0
TestedNov 15, 2026
Runtimetransformers 5.22.1

Quality criteria

No explicit quality gate recorded. Review per-case evidence; full certification is pending.

PrecisionDeviceResultPlatformInference / load
FP32mpsAcceleratedPassedFastmacos · aarch64487 ms / 2605 ms
BF16mpsAcceleratedPassedFastmacos · aarch64647 ms / 3852 ms
FP16mpsAcceleratedPassedFastmacos · aarch64658 ms / 4157 ms
FP32cpuPassedSlowmacos · aarch646388 ms / 2097 ms
BF16cpuPassedSlowmacos · aarch646966 ms / 1685 ms
Test setup & sample details
Quality gate
No explicit quality gate recorded. Review the per-case quality evidence; full certification is pending.
Tested
2026-11-15T21:15:24.467448+00:00
Environment
{ "os": "macOS 27.2", "hardware": "Apple M3 Max", "cpuThreads": 4, "engine": "Transformers 5.22.1 / PyTorch 2.10.0" }
Runtime
transformers 5.22.1
Artifact revision
7bca5038823c0d0f14efe06439109354a10e014c
Fixture
{
  "file": "research/lfm3/base-completion-fixtures.json",
  "size": 267,
  "sha1": "30bb1eee6b45d542a81bcc346b916acab19963fd",
  "sha256": "69694e5c4aee2a31ce0cc92b48444d0d8f1866426d4b36afa413ac5f57f593f9"
}
Rerun when
Artifact/runtime revisions; Generation/preprocessing changes; Fixture/quality scope; OS/backend behavior
Per-case quality & failure details
[
  {
    "precision": "fp32",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "metric": "expectedPrefixMatch",
      "score": 1,
      "passed": true
    },
    "runtime": {
      "artifactRevision": "7bca5038823c0d0f14efe06439109354a10e014c",
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 2096.693500003312,
      "inferenceMs": 6387.967666960321,
      "totalMs": 8484.661166963633,
      "tokensPerSecond": 1.8785317374203512,
      "grade": "slow"
    }
  },
  {
    "precision": "fp32",
    "device": "mps",
    "status": "pass",
    "quality": {
      "metric": "expectedPrefixMatch",
      "score": 1,
      "passed": true
    },
    "runtime": {
      "artifactRevision": "7bca5038823c0d0f14efe06439109354a10e014c",
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 2604.95070804609,
      "inferenceMs": 486.7613749811426,
      "totalMs": 3091.7120830272324,
      "tokensPerSecond": 24.652736673005098,
      "grade": "fast"
    }
  },
  {
    "precision": "bf16",
    "device": "cpu",
    "status": "pass",
    "quality": {
      "metric": "expectedPrefixMatch",
      "score": 1,
      "passed": true
    },
    "runtime": {
      "artifactRevision": "7bca5038823c0d0f14efe06439109354a10e014c",
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 1685.425584029872,
      "inferenceMs": 6965.816999960225,
      "totalMs": 8651.242583990097,
      "tokensPerSecond": 1.7226981415200142,
      "grade": "slow"
    }
  },
  {
    "precision": "bf16",
    "device": "mps",
    "status": "pass",
    "quality": {
      "metric": "expectedPrefixMatch",
      "score": 1,
      "passed": true
    },
    "runtime": {
      "artifactRevision": "7bca5038823c0d0f14efe06439109354a10e014c",
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 3852.06633299822,
      "inferenceMs": 646.5762499719858,
      "totalMs": 4498.642582970206,
      "tokensPerSecond": 18.559295984843125,
      "grade": "fast"
    }
  },
  {
    "precision": "fp16",
    "device": "mps",
    "status": "pass",
    "quality": {
      "metric": "expectedPrefixMatch",
      "score": 1,
      "passed": true
    },
    "runtime": {
      "artifactRevision": "7bca5038823c0d0f14efe06439109354a10e014c",
      "engine": "transformers",
      "version": "5.22.1",
      "torchVersion": "2.10.0",
      "loadMs": 4156.570166989695,
      "inferenceMs": 658.4046250209212,
      "totalMs": 4814.9747920106165,
      "tokensPerSecond": 18.225874399984193,
      "grade": "fast"
    }
  }
]
Release validation still required
  • Strict backup coverage, public dependency verification and cold model/interpreter/package-source recovery remain pending.

Something not quite right?

Tell us what happened with Liquid LFM3 Desktop. Loading, results, or the page itself—we’re here for all of it.

Attachments Optional

Up to 3 files · 50 MB each

Browser, OS, CPU/GPU, memory, screen, storage, and network details are included automatically. Model inputs, results, and browsing history are not collected.