Codegemma 7B — Hardware Requirements & GPU Compatibility
ChatCode
Codegemma 7B is a 8.5B-parameter open language model from Google in the Gemma family. At BF16 it needs about 18.78 GB of VRAM — see which GPUs and Macs can run it below.
2.0K downloads
220 likesSpecifications
- Publisher
- Family
- Gemma
- Parameters
- 8.5B
- Release Date
- 2024-08-07
- License
- Gemma Terms
Get Started
HuggingFace
How Much VRAM Does Codegemma 7B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 18.8 GB | — | 17.08 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Codegemma 7B?
BF16 · 18.8 GBCodegemma 7B (BF16) requires 18.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 25+ GB is recommended. 6 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroom
Which Devices Can Run Codegemma 7B?
BF16 · 18.8 GB21 devices with unified memory can run Codegemma 7B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Codegemma 7B need?
Codegemma 7B requires 18.8 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 8.5B × 16 bits ÷ 8 = 17.1 GB
KV Cache + Overhead ≈ 1.7 GB (at 2K context + ~0.3 GB framework)
VRAM usage by quantization
BF1618.8 GB- Can I run Codegemma 7B on a Mac?
Codegemma 7B requires at least 18.8 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Codegemma 7B locally?
Yes — Codegemma 7B can run locally on consumer hardware. At BF16 quantization it needs 18.8 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Codegemma 7B?
At BF16, Codegemma 7B can reach ~155 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~35 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 18.8 × 0.55 = ~155 tok/s
Estimated speed at BF16 (18.8 GB)
~155 tok/s~35 tok/s~116 tok/s~96 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Codegemma 7B?
At BF16, the download is about 17.08 GB.
- Which GPUs can run Codegemma 7B?
6 consumer GPUs can run Codegemma 7B at BF16 (18.8 GB). Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX. 1 GPU have plenty of headroom for comfortable inference.
- Which devices can run Codegemma 7B?
21 devices with unified memory can run Codegemma 7B at BF16 (18.8 GB), including Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB), Mac Pro M2 Ultra (192 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.