Vllm Translategemma 12B IT — Hardware Requirements & GPU Compatibility
ChatTranslation
Vllm Translategemma 12B IT is a 13.2B-parameter open language model from chbae624 in the Gemma family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 8.97 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q2_K · 16K)Best on 24 GB (Q4_K_M · ~73 tok/s · 128K)Macs from 32 GB (~9 tok/s)
Gemma Terms
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Vllm Translategemma 12B IT
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve chbae624/vllm-translategemma-12b-it --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (2 GB headroom) · Q4_K_M
- Generation speed
- ~26 tok/s
- generation speed
- Cost per 1M output tokens
- $0.80
- per 1M output tokens
How Much VRAM Does Vllm Translategemma 12B IT Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 6.7 GB | 8.7 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~98 tok/s | |
| 7.5 GB | 9.6 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~88 tok/s | |
| 9.0 GB | 11 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~73 tok/s | |
| 10.5 GB | 12.5 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~63 tok/s | |
| 11.9 GB | 14 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~55 tok/s | |
| 14.3 GB | 16.3 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~46 tok/s | |
| 27.4 GB | 29.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 28.2 GB |
Selected Q4_K_M
4-bit medium quantization9.8GBat 16K context
- Weights
- 7.9 GB
- KV cache
- 1.6 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 14.2 GB free · ~73 tok/s
≥50 tok/s needs ≥689 GB/s · RTX 5090 ~130, RTX 3090 Ti ~73
Which Hardware Can Run Vllm Translategemma 12B IT?
Q4_K_M · 9.0 GBVllm Translategemma 12B IT (Q4_K_M) requires 9.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 12+ GB is recommended. Using the full 131K context window can add up to 9.6 GB, bringing total usage to 18.6 GB. 40 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti. 155 devices with unified memory can run Vllm Translategemma 12B IT, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M2 Pro (16 GB).
12 GB
12 GB tier →fits · 2.2 GB free · up to 32K ctx
10 GPUs
RTX 3080 Ti~66Simulate output on RTX 3080 TiRTX 5070~49Simulate output on RTX 5070RTX 4070~37Simulate output on RTX 407020 GB
fits · 10.2 GB free · up to 128K ctx
1 GPU
RX 7900 XT~54Simulate output on RX 7900 XT32–40 GB
fits · 22.2 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~130Simulate output on RTX 5090M5 Max36 GB~36Simulate output on Mac Studio M5 Max (36 GB)M532 GB~12Simulate output on MacBook Air 13" M5 (32 GB)Orin~15Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~10Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Vllm Translategemma 12B IT
Gemma family · 16 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Mistral Nemo Base 2407Mistral AI | — | 11 GB | ~81 | same size |
| Phi 3 Medium 128k InstructMicrosoft | — | 12 GB | ~72 | same size |
| Qwen1.5 14BAlibaba | — | 22.2 GB | ~63 | same size |
| Baichuan2 13B Basebaichuan-inc | — | 11.5 GB | ~76 | same size |
| Llama 2 13B Chat HFMeta | — | 11.5 GB | ~76 | same size |
| Vicuna 13B V1.3LMSYS | — | 8.6 GB | ~76 | same size |
Specifications
- Publisher
- chbae624
- Family
- Gemma
- Parameters
- 13.2B
- Architecture
- Gemma3ForConditionalGeneration
- Context Length
- 131,072 tokens
- Vocabulary Size
- 262,208
- Release Date
- 2026-01-29
- License
- Gemma Terms
Frequently Asked Questions
How much VRAM does Vllm Translategemma 12B IT need?
Vllm Translategemma 12B IT requires 9.0 GB of VRAM at Q4_K_M, or 27.4 GB at BF16. Full 131K context adds up to 9.6 GB (18.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 13.2B × 4.8 bits ÷ 8 = 7.9 GB
KV Cache + Overhead ≈ 1.1 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 10.7 GB (at full 131K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Vllm Translategemma 12B IT?
Yes, at Q8_0 (14.3 GB) or lower. Higher quantizations like BF16 (27.4 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Vllm Translategemma 12B IT?
For Vllm Translategemma 12B IT, Q4_K_M (9.0 GB) offers the best balance of quality and VRAM usage. Q5_K_M (10.5 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 6.7 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Vllm Translategemma 12B IT on a Mac?
Yes, but only at lower quantizations. The smallest Mac that can run Vllm Translategemma 12B IT is Apple iPhone 17 Pro (12 GB) at Q2_K; 139 of the 156 Macs we list can run it at some quantization. For Q4_K_M (9.0 GB) you need a Mac with more unified memory.
Can I run Vllm Translategemma 12B IT locally?
Yes — Vllm Translategemma 12B IT can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Vllm Translategemma 12B IT?
At Q4_K_M, Vllm Translategemma 12B IT can reach ~535 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~73 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 8.97 × 0.65 = ~580 tok/s
Estimated speed at Q4_K_M (9.0 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Vllm Translategemma 12B IT?
At Q4_K_M, the download is about 7.92 GB. The full-precision BF16 version is 26.39 GB. The smallest option (Q2_K) is 5.61 GB.
Which GPUs can run Vllm Translategemma 12B IT?
40 consumer GPUs can run Vllm Translategemma 12B IT at Q4_K_M. Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
Which devices can run Vllm Translategemma 12B IT?
155 devices with unified memory can run Vllm Translategemma 12B IT at Q4_K_M (9.0 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.