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.

HY-MT1.5 1.8B — Translation

Translate short text locally in a browser using Tencent HY-MT1.5 1.8B and explicit language selections.

Browser inferencePowered by WebAI · Private · Unlimited runs
About this model

Translate short text locally in a browser using Tencent HY-MT1.5 1.8B and explicit language selections.

Created by
Tencent Hunyuan
Model released
Dec 30, 2025
Source repository created
Dec 25, 2025
Model weights · Q4F16
1.2 GB

Best for

  • Draft translations between provider-supported languages for human review.

Keep in mind

  • Translations can omit, invent or mistranslate facts, names, numbers, negation and idioms. Review important outputs.
  • Text-only translation; no automatic language detection, document layout preservation, image translation or streaming.
  • The adapter limits text to16384 characters and input-plus-output to4096 tokens as a resource policy, below the underlying model context.
  • Language choices are accepted selections, not evidence of measured quality in every pair.
  • No dedicated content safety filter; review generated translations for consequential uses.
  • Tencent HY Community License applies, including territory and use restrictions; this is not Apache/MIT licensed.
  • Large downloads and memory requirements; browser precision/device support is limited to measured passing combinations.
  • Five short authored cases establish limited compatibility, not broad language/domain quality.
  • Provider-supported language selections are enabled by explicit user exception. Only en→es, es→en, en→fr and zh→en have passing browser language evidence; other directions are unverified. Hebrew failed a retained diagnostic and remains unverified, not quality-certified.

Processed on your deviceInference runs in this browser. Input/output sharing requires consent for each run. Basic run logs include device information and approximate connection location. Model and runtime files download from external hosts. Saved downloads may be cleared by your browser.

Sources & technical details
Original model
tencent/HY-MT1.5-1.8B
Release verification
Verified
Release evidence
Original HY-MT1.5-1.8B release explicitly dated in upstream model-card news; excludes ONNX conversion.
Source updated
Jan 1, 2026
Browser model files
onnx-community/HY-MT1.5-1.8B-ONNX
Files updated
Jan 14, 2026
Artifact revision
2f11819b25de08cecd344735cdfa5136ade41a67

Input

TEXT

Sample result

HY-MT1.5 1.8B — Translation

Settings

Limits translation length; input plus requested output must fit the adapter4096-token budget. Lower values finish sooner but may truncate translation; higher values allow longer output and use more time/memory.

Output

Your translation will appear here

Enter English text or try the sample to translate it into Spanish.

Benchmark evidence

Saved worker evidence is separate from benchmarks run on your device. It does not guarantee speed or quality on every browser.

HY-MT1.5 1.8B — Translation

8 configurations tested
Passed2
Failed6
TestedNov 2, 2026
RuntimeTransformers.js 4.2.0

Quality criteria

{ "metric": "best-reference character-ngram F2 orders1–6", "minimumScore": 0.55, "requireSemanticAnchors": true, "rejectRepetitive": true, "scope": "All five authored cases in four accepted directions; Hebrew is rejected by validation after failing the retained broader diagnostic." }

PrecisionDeviceResult
Q4F16GPU · WebGPUAcceleratedPassedModerate
Q4GPU · WebGPUAcceleratedPassedModerate
Q4CPU · WASMFailed—
Q4F16CPU · WASMFailed—
FP16GPU · WebGPUFailed—
FP16CPU · WASMFailed—
FP32GPU · WebGPUFailed—
FP32CPU · WASMFailed—
Test setup & sample details
Quality gate
{ "metric": "best-reference character-ngram F2 orders1–6", "minimumScore": 0.55, "requireSemanticAnchors": true, "rejectRepetitive": true, "scope": "All five authored cases in four accepted directions; Hebrew is rejected by validation after failing the retained broader diagnostic." }
Tested
2026-11-02T15:19:01.046Z
Environment
{ "browser": "156.0.8078.79", "os": "Darwin", "release": "27.2.0", "arch": "arm64", "cpu": "Apple M3 Max", "runtime": { "name": "Transformers.js", "version": "4.2.0" }, "workerType": "module", "cache": "fresh disk-backed profile per case", "artifactTransport": "checksum-verified pinned local mirror", "wasmThreads": 1 }
Runtime
Transformers.js 4.2.0
Artifact revision
2f11819b25de08cecd344735cdfa5136ade41a67
Fixture
{
  "file": "model-samples/hunyuan-translation-browser.json",
  "url": "https://assets.localai.tools/samples/v1/files/858fa49a0d90bb76dcec9f86febec9536ff304d28321361f78239a153b011d8c/hunyuan-translation-browser.json",
  "size": 2810,
  "sha1": "b2b4c977b0cf11da74177cb838e7aa9258092bad",
  "sha256": "858fa49a0d90bb76dcec9f86febec9536ff304d28321361f78239a153b011d8c"
}
Rerun when
model artifact revision change; runtime version change; fixture or quality-gate change; backend/platform behavior change
Per-case quality & failure details
[
  {
    "precision": "q4",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "meanCharacterF2": 1,
      "cases": [
        {
          "fixtureId": "standard",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "negation",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "reverse",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "french",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "chinese",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        }
      ]
    },
    "runtime": {
      "loadMs": 3554.7999999970198,
      "inferenceMs": 1367.5999999940395,
      "totalMs": 6446.70000000298,
      "grade": "excellent",
      "charactersPerSecond": 135.2734717759625
    }
  },
  {
    "precision": "q4",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 5751.70000000298,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 9, ERROR_MESSAGE: Could not find an implementation for GatherBlockQuantized(1) node with name '/model/embed_tokens/Gather_Quant'"
  },
  {
    "precision": "q4f16",
    "device": "webgpu",
    "status": "pass",
    "quality": {
      "passed": true,
      "meanCharacterF2": 1,
      "cases": [
        {
          "fixtureId": "standard",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "negation",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "reverse",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "french",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        },
        {
          "fixtureId": "chinese",
          "passed": true,
          "characterF2": 1,
          "threshold": 0.55,
          "anchorsPassed": true,
          "repetitive": false,
          "metric": "best authored reference character-ngram F2 orders1–6"
        }
      ]
    },
    "runtime": {
      "loadMs": 5026.20000000298,
      "inferenceMs": 1328.3999999910593,
      "totalMs": 7622.600000008941,
      "grade": "excellent",
      "charactersPerSecond": 139.26528154264162
    }
  },
  {
    "precision": "q4f16",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 5566.4000000059605,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 9, ERROR_MESSAGE: Failed to find kernel for com.microsoft.GatherBlockQuantized(1) (node:'/model/embed_tokens/Gather_Quant' ep:'CPUExecutionProvider'). Kernel not found"
  },
  {
    "precision": "fp16",
    "device": "webgpu",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 6888.399999991059,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: Deserialize tensor model.layers.22.attn.o_proj.MatMul.weight failed.Failed to load external data file \"\"model_fp16.onnx_data_1\"\", error: File not found in preloaded files."
  },
  {
    "precision": "fp16",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 7292.699999988079,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: /mnt/vss/_work/1/s/onnxruntime/core/optimizer/initializer.cc:51 onnxruntime::Initializer::Initializer(const onnx::TensorProto &, const std::filesystem::path &) [ONNXRuntimeError] : 1 : FAIL : Failed to load external data file \"\"model_fp16.onnx_data_1\"\", error: File not found in preloaded files.\n"
  },
  {
    "precision": "fp32",
    "device": "webgpu",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 7060.300000011921,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: Deserialize tensor model.layers.6.mlp.down_proj.MatMul.weight failed.Failed to load external data file \"\"model.onnx_data_1\"\", error: File not found in preloaded files."
  },
  {
    "precision": "fp32",
    "device": "wasm",
    "status": "fail",
    "quality": "Not embedded",
    "runtime": {
      "loadMs": null,
      "inferenceMs": null,
      "totalMs": 6775.9000000059605,
      "grade": "failed"
    },
    "failure": "Can't create a session. ERROR_CODE: 1, ERROR_MESSAGE: Deserialize tensor model.layers.11.mlp.gate_proj.MatMul.weight failed.Failed to load external data file \"\"model.onnx_data_1\"\", error: File not found in preloaded files."
  }
]

Something not quite right?

Tell us what happened with HY-MT 1.5. 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.