Mark 1x 9B — Hardware Requirements & GPU Compatibility
ChatFunctions
Mark 1x 9B is a 9.0B-parameter open language model from Saanora. It supports a context window of up to 262,144 tokens. At BF16 it needs about 18.48 GB of VRAM — see which GPUs and Macs can run it below.
Runs well on a 24 GB GPU
Runs on 24 GB (BF16 · ~36 tok/s · 128K)Macs from 24 GB (BF16 · ~12 tok/s)
Other
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Mark 1x 9B
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve Saanora/mark-1x-9b --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3090 (24 GB) (5 GB headroom) · BF16
- Generation speed
- ~33 tok/s
- generation speed
- Cost per 1M output tokens
- $1.43
- per 1M output tokens
How Much VRAM Does Mark 1x 9B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 18.5 GB | 19.3 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 128K context · ~36 tok/s |
Selected BF16
Brain floating point 1618.8GBat 16K context
- Weights
- 17.9 GB
- KV cache
- 0.5 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 5.3 GB free · ~36 tok/s
≥50 tok/s needs ≥1.42 TB/s · RTX 5090 ~63
Which Hardware Can Run Mark 1x 9B?
BF16 · 18.5 GBMark 1x 9B (BF16) requires 18.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 25+ GB is recommended. Using the full 262K context window can add up to 8.3 GB, bringing total usage to 26.8 GB. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Mark 1x 9B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Mini M5 Pro (24 GB).
20 GB
fits · 1.3 GB free · up to 32K ctx
1 GPU
RX 7900 XT~26Simulate output on RX 7900 XT32–40 GB
fits · 13.3 GB free · up to 256K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~63Simulate output on RTX 5090M5 Max36 GB~17Simulate output on Mac Studio M5 Max (36 GB)M532 GB~6Simulate output on MacBook Air 13" M5 (32 GB)Orin~7Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~5Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at BF16
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Qwen3.5 9BAlibaba | 99.4 | 6.6 GB | ~103 | same size |
| Granite Switch 4.1 8BIBM | — | 8.4 GB | ~111 | same size |
| Ornith 1.5 9Bornith-ai | — | 6.6 GB | ~103 | same size |
| Apertus V1.5 8Bswiss-ai | — | 8.5 GB | ~111 | same size |
| LFM2.5 8B A1BLiquid AI | — | 5.6 GB | ~335 | same size |
| MiniCPM O 4 5OpenBMB | — | 8.3 GB | ~105 | same size |
Specifications
- Publisher
- Saanora
- Parameters
- 9.0B
- Architecture
- Qwen3_5ForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-09-08
- License
- Other
Frequently Asked Questions
How much VRAM does Mark 1x 9B need?
Mark 1x 9B requires 18.5 GB of VRAM at BF16. Full 262K context adds up to 8.3 GB (26.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 9.0B × 16 bits ÷ 8 = 17.9 GB
KV Cache + Overhead ≈ 0.6 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 ≈ 8.9 GB (at full 262K context)
VRAM usage by quantization
Can I run Mark 1x 9B on a Mac?
Yes — Mac Mini M5 Pro (24 GB) and 106 other Macs can run Mark 1x 9B. Apple Silicon uses unified memory, so the model shares RAM with the system. At BF16 you need at least 18.5 GB of usable unified memory (RAM minus macOS overhead).
Can I run Mark 1x 9B locally?
Yes — Mark 1x 9B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Mark 1x 9B?
At BF16, Mark 1x 9B can reach ~260 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~36 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 ÷ 18.5 × 0.65 = ~281 tok/s
Estimated speed at BF16 (18.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Mark 1x 9B?
At BF16, the download is about 17.91 GB.
Which GPUs can run Mark 1x 9B?
8 consumer GPUs can run Mark 1x 9B at BF16. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX. 1 GPU has plenty of headroom for comfortable inference.
Which devices can run Mark 1x 9B?
122 devices with unified memory can run Mark 1x 9B at BF16 (18.5 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.