T5gemma B B Ul2 IT — Hardware Requirements & GPU Compatibility
Chat
T5gemma B B Ul2 IT is a 591M-parameter open language model from Google in the Gemma family. At BF16 it needs about 1.30 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (BF16 · 32K)Best on 24 GB (BF16 · ~504 tok/s · 32K)Macs from 8 GB (BF16 · ~55 tok/s)
Gemma Terms
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run T5gemma B B Ul2 IT
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve google/t5gemma-b-b-ul2-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 (9 GB headroom) · BF16
- Generation speed
- ~180 tok/s
- generation speed
- Cost per 1M output tokens
- $0.13
- per 1M output tokens
How Much VRAM Does T5gemma B B Ul2 IT Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.3 GB | 4 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 32K context · ~504 tok/s |
Selected BF16
Brain floating point 162.5GBat 16K context
- Weights
- 0.8 GB
- KV cache
- 1.4 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 21.5 GB free · ~504 tok/s
≥50 tok/s needs ≥100 GB/s · RTX 5090 ~896, RTX 3090 Ti ~504
Which Hardware Can Run T5gemma B B Ul2 IT?
BF16 · 1.3 GBT5gemma B B Ul2 IT (BF16) requires 1.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run T5gemma B B Ul2 IT, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 5.5 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~380Simulate output on RTX 3080RTX 2080 Ti~308Simulate output on RTX 2080 TiRTX 3070 Ti~304Simulate output on RTX 3070 TiM3*8 GB~55Simulate output on iMac 24" M3 (8 GB)Orin Nano~51Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~37Simulate output on iPhone 1720 GB
fits · 17.5 GB free · up to 32K ctx
1 GPU
RX 7900 XT~369Simulate output on RX 7900 XT32–40 GB
fits · 29.5 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~896Simulate output on RTX 5090M5 Max36 GB~248Simulate output on Mac Studio M5 Max (36 GB)M532 GB~83Simulate output on MacBook Air 13" M5 (32 GB)Orin~102Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~68Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at BF16
Alternatives to T5gemma B B Ul2 IT
Gemma family · 16 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| D1 Omni 600MLiquid AI | — | 1.2 GB | ~910 | same size |
| Qwen3 0.6B BaseAlibaba | — | 1.6 GB | ~840 | same size |
| Qwen1.5 0.5B ChatAlibaba | — | 2.3 GB | ~753 | same size |
| Bloomz 560MBigScience | — | 0.4 GB | ~1771 | same size |
| Bloom 560MBigScience | — | 1.6 GB | ~1771 | same size |
| Xglm 564MMeta | — | 0.4 GB | ~1771 | same size |
Specifications
- Publisher
- Family
- Gemma
- Parameters
- 591M
- Release Date
- 2025-06-19
- License
- Gemma Terms
Frequently Asked Questions
How much VRAM does T5gemma B B Ul2 IT need?
T5gemma B B Ul2 IT requires 1.3 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 591M × 16 bits ÷ 8 = 1.2 GB
KV Cache + Overhead ≈ 0.1 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.
VRAM usage by quantization
Can I run T5gemma B B Ul2 IT on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run T5gemma B B Ul2 IT. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 1.3 GB of usable unified memory (RAM minus macOS overhead).
Can I run T5gemma B B Ul2 IT locally?
Yes — T5gemma B B Ul2 IT can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is T5gemma B B Ul2 IT?
At BF16, T5gemma B B Ul2 IT can reach ~3692 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~504 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 ÷ 1.3 × 0.65 = ~4000 tok/s
Estimated speed at BF16 (1.3 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of T5gemma B B Ul2 IT?
At BF16, the download is about 1.18 GB.
Which GPUs can run T5gemma B B Ul2 IT?
52 consumer GPUs can run T5gemma B B Ul2 IT at BF16. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run T5gemma B B Ul2 IT?
163 devices with unified memory can run T5gemma B B Ul2 IT at BF16 (1.3 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.