Academic Ds 9B — Hardware Requirements & GPU Compatibility
Chat
Academic Ds 9B is a 9.4B-parameter open language model from ByteDance-Seed. It supports a context window of up to 8,192 tokens. At Q4_K_M it needs about 6.19 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 8K)Best on 24 GB (Q4_K_M · ~399 tok/s · 8K)Macs from 8 GB (Q2_K · ~71 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Academic Ds 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 ByteDance-Seed/academic-ds-9B --max-model-len 8192vLLM repo with its own quantisation
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (5 GB headroom) · Q4_K_M
- Generation speed
- ~184 tok/s
- generation speed
- Cost per 1M output tokens
- $0.093
- per 1M output tokens
How Much VRAM Does Academic Ds 9B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 4.5 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 8K context · ~481 tok/s | |
| 5.1 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 8K context · ~448 tok/s | |
| 6.2 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 8K context · ~399 tok/s | |
| 7.2 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 8K context · ~359 tok/s | |
| 8.3 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 8K context · ~326 tok/s | |
| 9.9 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 8K context · ~286 tok/s | |
| 19.3 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 8K context · ~168 tok/s |
Selected Q4_K_M
4-bit medium quantization6.1GBat 8K context
- Weights
- 5.6 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 17.9 GB free · ~399 tok/s
≥50 tok/s needs ≥86 GB/s · RTX 5090 ~557, RTX 3090 Ti ~399
Which Hardware Can Run Academic Ds 9B?
Q4_K_M · 6.2 GBAcademic Ds 9B (Q4_K_M) requires 6.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 9+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti. 151 devices with unified memory can run Academic Ds 9B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA Jetson Orin Nano 8GB (Super).
≤ 11 GB
≤ 11 GB tier →fits · 1.9 GB free · up to 8K ctx
16 GPUs · 1 device family
RTX 3080~329Simulate output on RTX 3080RTX 2080 Ti~282Simulate output on RTX 2080 TiRTX 3070 Ti~279Simulate output on RTX 3070 TiOrin Nano~59Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)20 GB
fits · 13.9 GB free · up to 8K ctx
1 GPU
RX 7900 XT~184Simulate output on RX 7900 XT32–40 GB
fits · 25.9 GB free · up to 8K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~557Simulate output on RTX 5090M5 Max36 GB~154Simulate output on Mac Studio M5 Max (36 GB)M532 GB~76Simulate output on MacBook Air 13" M5 (32 GB)Orin~112Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~73Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
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 |
| MiniCPM O 4 5OpenBMB | — | 8.3 GB | ~105 | same size |
| Qwen3 VL 8B InstructAlibaba | — | 8 GB | ~112 | same size |
Specifications
- Publisher
- ByteDance-Seed
- Parameters
- 9.4B
- Active parameters
- 1.6B per token
- Architecture
- DeepseekV3ForCausalLM
- Context Length
- 8,192 tokens
- Vocabulary Size
- 129,280
- Release Date
- 2025-04-09
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Academic Ds 9B need?
Academic Ds 9B requires 6.2 GB of VRAM at Q4_K_M, or 19.3 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 9.4B × 4.8 bits ÷ 8 = 5.6 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 ≈ 0.5 GB (at full 8K context)
VRAM usage by quantization
What's the best quantization for Academic Ds 9B?
For Academic Ds 9B, Q4_K_M (6.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (7.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 4.5 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Academic Ds 9B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Academic Ds 9B, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 6.2 GB of usable unified memory (RAM minus macOS overhead), from Mac Mini M2 Pro (16 GB).
Can I run Academic Ds 9B locally?
Yes — Academic Ds 9B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Academic Ds 9B?
At Q4_K_M, Academic Ds 9B can reach ~292 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~399 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 1.1 GB active ÷ (8000 × 0.65) = 0.2053 ms, plus 16 layers × 0.055 ms = 0.8800 ms, so 1000 ÷ 1.0853 ms = ~921 tok/s
Estimated speed at Q4_K_M (6.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Academic Ds 9B?
At Q4_K_M, the download is about 5.62 GB. The full-precision BF16 version is 18.73 GB. The smallest option (Q2_K) is 3.98 GB.
Which GPUs can run Academic Ds 9B?
52 consumer GPUs can run Academic Ds 9B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 40 GPUs have plenty of headroom for comfortable inference.
Which devices can run Academic Ds 9B?
151 devices with unified memory can run Academic Ds 9B at Q4_K_M (6.2 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.