tencent
Tencent: Hy-MT2-1.8B
tencent/hy-mt2-1.8b
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
tencent/hy-mt2-1.8b is a compact 1.8-billion-parameter text-to-text translation model from Tencent's Hy-MT2 family, described by its documentation as covering 33 languages plus several Chinese dialect and minority-language pairs. It is designed for instruction-guided translation workflows, including terminology control, style guidance, delimiter preservation, contextual background, and structured-data formats. Its catalog listing reports an 8192-token context window. Vendor-published evaluations claim it outperforms some mainstream commercial translation APIs overall, but these are self-reported results rather than independent measurements.[1][2]
Strengths
- Compact footprint: Tencent reports a 1.25-bit quantized variant needing roughly 440 MB of storage with about a 1.5x inference speedup, aimed at on-device deployment.[2]
- Broad instruction-following surface: documented prompt templates support terminology references, style constraints, personalization preferences, delimiter preservation, and structured data such as JSON.[1][2]
- Wide language coverage for its size, spanning 33 listed languages including lower-resource ones such as Tibetan, Uyghur, and Cantonese.[2]
Limitations
- Evidence is vendor and catalog only: no independent benchmark figures were in the source pack, performance claims are self-reported, and quality should be validated on your own workloads before adoption.[1][2]
- Text-only input and output with no advertised reasoning mode, an 8192-token context cap, and no default system prompt; the vendor recommends specific decoding settings for best results.[1][2]
- Integration friction: documentation requires transformers 5.6.0 or newer with trust_remote_code, and vLLM or SGLang must be built from source rather than installed as released packages.[2]
Where to use it
- Cost-sensitive, high-volume multilingual translation pipelines where a small dedicated model may be preferable to a general-purpose LLM.[1][2]
- On-device or edge translation features using the documented quantized GGUF variants, subject to verifying quality on the target language pairs.[2]
- Localization workflows needing formatting guarantees, such as preserving placeholders and keys in structured files or applying terminology glossaries and style rules.[1][2]
Sources
- [1] OpenRouter catalog · Tencent: Hy-MT2-1.8BSources checked: Nov 13, 2026
- [2] Model documentation · Tencent: Hy-MT2-1.8BSources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided...
Specifications
- Model ID
- tencent/hy-mt2-1.8b
- Canonical ID
- tencent/hy-mt2-1.8b-20260521
- Creator
- tencent
- Added to OpenRouter
- 20.08.2026
- Context length
- 8,192
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.044 / out $0.177
- Knowledge cutoff
- —
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | — |
| Coding | No measurement | — |
| Agents | No measurement | — |
| Writing | No measurement | output: text |
| Reasoning | No measurement | — |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| Tencentfp8 | 8,192 | $0.044 | $0.177 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.