Apertus 8B Instruct 2509 — Hardware Requirements & GPU Compatibility
Chat
Apertus 8B Instruct is an open-source instruction-tuned model from Swiss AI, a collaborative research initiative. Built on an 8 billion parameter base, it emphasizes transparency, open data, and European AI sovereignty. For local users, it delivers solid general-purpose chat and instruction-following in a standard 8B footprint that runs well on consumer GPUs with 8 to 10 GB of VRAM, making it a practical choice for those who value open, community-driven model development.
117.9K downloads
439 likesSpecifications
- Publisher
- swiss-ai
- Family
- Apertus
- Parameters
- 8B
- Architecture
- ApertusForCausalLM
- Context Length
- 65,536 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2025-08-13
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does Apertus 8B Instruct 2509 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 4.0 GB | 12.3 GB | 3.40 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 4.1 GB | 12.4 GB | 3.50 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 4.5 GB | 12.8 GB | 3.90 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 4.6 GB | 12.9 GB | 4.00 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 5.4 GB | 13.7 GB | 4.80 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 6.3 GB | 14.6 GB | 5.70 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 7.2 GB | 15.5 GB | 6.60 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 8.6 GB | 16.9 GB | 8.00 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Apertus 8B Instruct 2509?
Q4_K_M · 5.4 GBApertus 8B Instruct 2509 (Q4_K_M) requires 5.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 7+ GB is recommended. Using the full 66K context window can add up to 8.3 GB, bringing total usage to 13.7 GB. 50 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Apertus 8B Instruct 2509?
Q4_K_M · 5.4 GB58 devices with unified memory can run Apertus 8B Instruct 2509, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroomWhere to Download Apertus 8B Instruct 2509
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does Apertus 8B Instruct 2509 need?
Apertus 8B Instruct 2509 requires 5.4 GB of VRAM at Q4_K_M, or 16.6 GB at BF16. Full 66K context adds up to 8.3 GB (13.7 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 8B × 4.8 bits ÷ 8 = 4.8 GB
KV Cache + Overhead ≈ 0.6 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 8.9 GB (at full 66K context)
VRAM usage by quantization
Q4_K_M5.4 GBQ4_K_M + full context13.7 GB- What's the best quantization for Apertus 8B Instruct 2509?
For Apertus 8B Instruct 2509, Q4_K_M (5.4 GB) offers the best balance of quality and VRAM usage. Q5_0 (5.6 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 2.8 GB.
VRAM requirement by quantization
IQ2_XXS2.8 GBQ2_K4.0 GBIQ4_XS4.9 GBQ4_K_M ★5.4 GBQ5_05.6 GBBF1616.6 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Apertus 8B Instruct 2509 on a Mac?
Apertus 8B Instruct 2509 requires at least 2.8 GB at IQ2_XXS, 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 Apertus 8B Instruct 2509 locally?
Yes — Apertus 8B Instruct 2509 can run locally on consumer hardware. At Q4_K_M quantization it needs 5.4 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Apertus 8B Instruct 2509?
At Q4_K_M, Apertus 8B Instruct 2509 can reach ~819 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~122 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 ÷ 5.4 × 0.65 = ~968 tok/s
Estimated speed at Q4_K_M (5.4 GB)
~968 tok/s~122 tok/s~968 tok/s~819 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Apertus 8B Instruct 2509?
At Q4_K_M, the download is about 4.80 GB. The full-precision BF16 version is 16.00 GB. The smallest option (IQ2_XXS) is 2.20 GB.
- Which GPUs can run Apertus 8B Instruct 2509?
50 consumer GPUs can run Apertus 8B Instruct 2509 at Q4_K_M (5.4 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 39 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Apertus 8B Instruct 2509?
59 devices with unified memory can run Apertus 8B Instruct 2509 at Q4_K_M (5.4 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, 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.