Hy-MT2 7B — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
The mid-size Hy-MT2 translation model, 33 languages. Tencent reports it ahead of much larger general models such as DeepSeek-V4-Pro in fast, non-thinking translation (vendor-reported). Official Q4_K_M GGUF 4.6 GB; Apache-2.0.
Hy-MT2 7B needs about 5 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Includes the KV cache at 8K context, so it reads higher than the headline figure.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 4.5 GB | ~155 tok/s | Fits | |
| Q3_K_M 3.41 bpw | 5.3 GB | ~134 tok/s | Fits | |
| Q4_K_M 4.83 bpw | 6.5 GB | ~107 tok/s | Fits | |
| Q5_K_M 5.67 bpw | 7.6 GB | ~95 tok/s | Fits | |
| Q6_K 6.56 bpw | 8.5 GB | ~86 tok/s | Fits | |
| Q8_0 8.50 bpw | 9.9 GB | ~70 tok/s | Fits | |
| F16 16.00 bpw | 17.9 GB | ~41 tok/s | Fits |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. Hy-MT2 7B VRAM calculator →
Get Hy-MT2 7B running
The cheapest catalogued GPU that runs Hy-MT2 7B is the Intel Arc B570 (10 GB).
or compare on Vast.ai from $0.35/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run Hy-MT2 7B
Install Ollama, then run:
ollama run hy-mtSpecifications
Verified — Checked against the primary source — the model card or the vendor spec page — and corroborated by a second independent source.
- Parameters
- 7 Billion (8.0B in the files)
- Context window
- 262,144
- Architecture
- Dense transformer (GQA)
- Provider
- Tencent
- Licence
- Apache-2.0
- Availability
- Weights downloadable
- Practical deployment
- Laptop
- Specified at
- Q4_K_M
- System RAM
- 16 GB
- Record updated
- 2026-11-11
Commercial use permitted. No usage restrictions beyond attribution.
See also Hy-MT2 30B-A3B
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Other Hy-MT2 sizes
Hy-MT2 7B — frequently asked questions
How much VRAM does Hy-MT2 7B need?
About 5 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Hy-MT2 7B run on an RTX 4090 (24 GB)?
Yes. Hy-MT2 7B needs about 5 GB at Q4_K_M, inside a 24 GB card, at an estimated 107 tokens/sec.
How do I run Hy-MT2 7B locally?
Install Ollama and run `ollama run hy-mt`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.
What other sizes does Hy-MT2 come in?
Hy-MT2 1.8B (2 GB), Hy-MT2 7B (5 GB), Hy-MT2 30B-A3B (19 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.