Hardware to run Qwen 3.5 9B

Qwen 3.5 9B (Mar 2 2026). Gated DeltaNet hybrid-attention multimodal vision-language model. 262K native context (up to 1M via YaRN), 201 languages, Apache 2.0. MMLU-Pro 82.5, GPQA Diamond 81.7, MathVision 78.9. Fits a 16GB laptop at Q4 (~5.7 GB). Vision: MMBench-EN 90.1, Video-MME 84.5.

Quantization Q2 Q2_K Q4 Q4_K_M Q5 Q5_K_M Q8 Q8_0

Availability ⚡ Only plug & play

Cheapest

Single Tesla P100 16 GB (used) build

NVIDIA · desktop tower

$500

tokens / secQ4

8B 33 t/s

14B 19 t/s

30B —

Memory16 GB · 15 usable

Bandwidth732 GB/s

Idle / Active25 W / 250 W

Sticker$500

Why: Lowest sticker that still fits Qwen 3.5 9B ($500 USD).

AmazoneBay (used)Newegg

📺 Reviews on YouTube

▶ Can a 10-Year-Old $5,700 GPU Beat a New $430 GPU? | Tesla P100 Local AI Review

Fastest

DGX B200 — 8× B200 server (1.44 TB HBM3e)

NVIDIA · 10U DGX server

$475,000

tokens / secQ4

8B 600 t/s

14B 420 t/s

30B 270 t/s

Memory1440 GB · 1404 usable

Bandwidth8000 GB/s

Idle / Active900 W / 10200 W

Sticker$475,000

Why: Highest measured tg/s — 533 t/s on Qwen 3.5 9B-class models at Q4.

NVIDIA DGX B200SuperMicro HGX B200Lambda Labs

📺 Reviews on YouTube

▶ Inside a 1.44 TB HBM3e NVIDIA HGX B200 AI Server from ASRock Rack

All-rounder

Tesla V100 32 GB SXM2 mod build

NVIDIA · desktop tower

$900

tokens / secQ4

8B 85 t/s

14B 50 t/s

30B 27 t/s

Memory32 GB · 31 usable

Bandwidth900 GB/s

Idle / Active33 W / 300 W

Sticker$900

Why: Top quartile across speed, value, memory headroom, and efficiency — the "buy this if unsure" pick.

AmazoneBay V100 SXM2 32 GBeBay SXM2 → PCIe adapter

📺 Reviews on YouTube

▶ Expensive RTX 5090 for LLMs? No — use V100 SXM2 + Z8 G4 instead ▶ I built my own AI — A.I.D.E.N. (32 GB V100 SXM2 + HP Z8 G4)

Best value

Single AMD Instinct MI50 32 GB (used) build

AMD · desktop tower

$900

tokens / secQ4

8B 71 t/s

14B 38 t/s

30B —

Memory32 GB · 31 usable

Bandwidth1024 GB/s

Idle / Active18 W / 300 W

Sticker$900

Why: Best $/tg-per-second — ~$14 per t/s.

AmazoneBay (used)Alibaba

📺 Reviews on YouTube

▶ AMD MI50 32 GB speed test — Ollama vs llama.cpp (GPT-OSS & Qwen3) ▶ Is the Radeon Instinct MI50 32 GB the ultimate cheap home-AI GPU?

Best CUDA

DGX H200 — 8× H200 server (1.13 TB HBM3e)

NVIDIA · 8U DGX / HGX server rack

$380,000

tokens / secQ4

8B 560 t/s

14B 390 t/s

30B 250 t/s

Memory1128 GB · 1100 usable

Bandwidth4800 GB/s

Idle / Active700 W / 6500 W

Sticker$380,000

Why: Strongest CUDA-only software stack among fitting builds.

NVIDIA DGX H200SuperMicro HGXLambda Labs

📺 Reviews on YouTube

▶ Inside the SUPER NVIDIA H200 Server From Supermicro (8U HGX H200)

Most VRAM

12× RTX Pro 6000 Blackwell rack (1152 GB)

NVIDIA · 8U server rack (multi-node, 1-2 chassis)

$118,000

tokens / secQ4

8B 460 t/s

14B 340 t/s

30B 250 t/s

Memory1152 GB · 1116 usable

Bandwidth1792 GB/s

Idle / Active340 W / 7400 W

Sticker$118,000

Why: 1116 GB usable — most headroom for batching and longer contexts.

AmazonCoreWeaveSuperMicroSHI enterprise

Efficient

MacBook Air M4 (16 GB)

Apple · laptop

$1,099

tokens / secQ4

8B 18 t/s

14B 8.0 t/s

30B —

Memory16 GB · 11 usable

Bandwidth120 GB/s

Idle / Active5 W / 30 W

Sticker$1,099

Why: 30 W active — lowest power draw of the fitting builds.

AmazonAppleB&H Photo

📺 Reviews on YouTube

▶ The budget MacBook so stubborn it survived a 44k-token test

Cheapest

Mac Mini M4 (16 GB)

Apple · mini desktop

$799

tokens / secQ4

8B 22 t/s

14B 10 t/s

30B —

Memory16 GB · 11 usable

Bandwidth120 GB/s

Idle / Active4 W / 50 W

Sticker$799

Why: Lowest sticker that still fits Qwen 3.5 9B ($799 USD).

AmazonAppleB&H Photo

📺 Reviews on YouTube

▶ Mac Mini M4 vs M3 Pro, 6700XT, 3080Ti — LLM Ollama side by side ▶ Intel Macbook vs Apple Silicon (M1, M3 Pro, M4) running LLM Ollama ▶ Ollama Mac MLX is here — 2× faster t/s for Apple Silicon

Fastest

Single AMD Instinct MI355X 288 GB workstation

AMD · 4U server (OAM, liquid-cooled)

$28,000

tokens / secQ4

8B 420 t/s

14B 270 t/s

30B 160 t/s

Memory288 GB · 282 usable

Bandwidth8000 GB/s

Idle / Active140 W / 1400 W

Sticker$28,000

Why: Highest measured tg/s — 373 t/s on Qwen 3.5 9B-class models at Q4.

Dell XE9712AMD MI350 seriesSupermicro UBB

📺 Reviews on YouTube

▶ AMD's MI350/355X Advancing AI Event Recap ▶ AMD AI Event w/ Craft Computing — MI350/355X Launch, ROCm 7 ▶ AMD Visit and Tour! Featuring ROCm 7 and AMD Instinct

All-rounder

Mac Studio M4 Max 128 GB

Apple · small desktop

$4,699

tokens / secQ4

8B 95 t/s

14B 55 t/s

30B 30 t/s

Memory128 GB · 112 usable

Bandwidth546 GB/s

Idle / Active8 W / 130 W

Sticker$4,699

Why: Top quartile across speed, value, memory headroom, and efficiency — the "buy this if unsure" pick.

AppleB&H PhotoAdorama

📺 Reviews on YouTube

▶ M3 Ultra vs M4 Max — Don't Buy the WRONG Mac Studio ▶ M5 MacBook Pro vs M4 Max & M3 Ultra for Local AI

Best value

Single AMD Radeon RX 9070 XT 16 GB build

AMD · desktop tower

$1,300

tokens / secQ4

8B 78 t/s

14B 44 t/s

30B —

Memory16 GB · 15 usable

Bandwidth645 GB/s

Idle / Active17 W / 304 W

Sticker$1,300

Why: Best $/tg-per-second — ~$19 per t/s.

AmazonNeweggB&H PhotoAMD

📺 Reviews on YouTube

▶ AMD 9070 XT FULL SEND — Ollama / llama.cpp / vLLM, gpt-oss / qwen3 / trinity / devstral all tested ▶ RX 9070 XT is cheaper AND better… mostly

Best CUDA

Single B200 180 GB workstation

NVIDIA · workstation / 4U server

$47,000

tokens / secQ4

8B 360 t/s

14B 225 t/s

30B 135 t/s

Memory180 GB · 176 usable

Bandwidth8000 GB/s

Idle / Active100 W / 1000 W

Sticker$47,000

Why: Strongest CUDA-only software stack among fitting builds.

NVIDIA B200 partnersSHI enterpriseLambda Labs

📺 Reviews on YouTube

▶ AI Lab: NVIDIA B200 vs GB200 explained | GPU architecture for LLMs ▶ Inside a NEW AI Cluster — Tour with NVIDIA B200

Most VRAM

Mac Studio M3 Ultra 512 GB

Apple · small desktop

$14,199

tokens / secQ4

8B 110 t/s

14B 70 t/s

30B 38 t/s

Memory512 GB · 480 usable

Bandwidth819 GB/s

Idle / Active12 W / 220 W

Sticker$14,199

Why: 480 GB usable — most headroom for batching and longer contexts.

AppleB&H Photo

📺 Reviews on YouTube

▶ M3 Ultra Mac Studio Review ▶ M3 Ultra vs RTX 5090 — The Final Battle

Efficient

MacBook Air M4 (16 GB)

Apple · laptop

$1,099

tokens / secQ4

8B 18 t/s

14B 8.0 t/s

30B —

Memory16 GB · 11 usable

Bandwidth120 GB/s

Idle / Active5 W / 30 W

Sticker$1,099

Why: 30 W active — lowest power draw of the fitting builds.

AmazonAppleB&H Photo

📺 Reviews on YouTube

▶ The budget MacBook so stubborn it survived a 44k-token test

Every other build that runs Qwen 3.5 9B

60 additional builds fit Qwen 3.5 9B at Q4_K_M (8 GB usable minimum), sorted by sticker price.

Build	Price	Memory	Bandwidth	tg/s (Q4)	Active W	5-yr power
Single Tesla P40 24 GB (used) buildNVIDIA · desktop tower	$750	24 / 23 GB	347 GB/s	40 t/s	250 W	$854
RTX 3060 12 GB buildNVIDIA · desktop tower	$900	12 / 11 GB	360 GB/s	34 t/s	170 W	$598
Mac Mini M4 (24 GB)Apple · mini desktop	$999	24 / 18 GB	120 GB/s	20 t/s	50 W	$177
Single Intel Arc B580 12 GB buildIntel · desktop tower	$1.1k	12 / 11 GB	456 GB/s	44 t/s	190 W	$664
MacBook Air M5 (16 GB)Apple · laptop	$1.3k	16 / 11 GB	153 GB/s	20 t/s	30 W	$115
Single RTX 3090 (used) buildNVIDIA · desktop tower	$1.5k	24 / 23 GB	936 GB/s	80 t/s	350 W	$1.2k
Single Intel Arc Pro B70 buildIntel · desktop tower	$1.8k	32 / 31 GB	608 GB/s	62 t/s	220 W	$782
Single AMD Radeon AI Pro R9700 32 GB buildAMD · desktop tower	$2.0k	32 / 31 GB	640 GB/s	68 t/s	300 W	$1.1k
Mac Studio M4 Max 36 GBApple · small desktop	$2.5k	36 / 28 GB	546 GB/s	84 t/s	130 W	$453
Quad AMD MI50 32 GB (128 GB) homelab buildAMD · rack/large tower	$2.5k	128 / 122 GB	1024 GB/s	64 t/s	1200 W	$4.2k
Quad Tesla P40 (96 GB) homelab buildNVIDIA · rack/large tower	$2.7k	96 / 92 GB	347 GB/s	43 t/s	1000 W	$3.5k
AMD Ryzen AI Max+ 395 (128 GB)AMD · mini desktop / laptop	$2.8k	128 / 96 GB	256 GB/s	44 t/s	120 W	$420
Dual RTX 3090 (used) buildNVIDIA · desktop tower	$2.8k	48 / 46 GB	936 GB/s	84 t/s	700 W	$2.4k
MacBook Pro M4 Pro 48 GBApple · laptop	$2.9k	48 / 40 GB	273 GB/s	44 t/s	70 W	$246
MacBook Pro M5 Pro 48 GBApple · laptop	$3.2k	48 / 40 GB	307 GB/s	53 t/s	75 W	$263
Single RTX 4090 buildNVIDIA · desktop tower	$3.2k	24 / 23 GB	1008 GB/s	116 t/s	410 W	$1.4k
Dual Intel Arc Pro B70 buildIntel · desktop tower	$3.2k	64 / 62 GB	608 GB/s	80 t/s	380 W	$1.4k
ASUS Ascent GX10 (128 GB)ASUS · small desktop	$3.5k	128 / 119 GB	273 GB/s	76 t/s	240 W	$903
Single AMD Radeon Pro W7800 32 GB buildAMD · workstation	$3.5k	32 / 31 GB	576 GB/s	52 t/s	260 W	$920
Dual AMD Radeon AI Pro R9700 build (64 GB)AMD · workstation	$3.7k	64 / 62 GB	640 GB/s	80 t/s	600 W	$2.1k
MacBook Pro M4 Max 64 GBApple · laptop	$4.0k	64 / 54 GB	410 GB/s	62 t/s	90 W	$315
Dell Pro Max with GB10 (128 GB)Dell · small desktop	$4.1k	128 / 119 GB	273 GB/s	80 t/s	240 W	$887
MacBook Pro M5 Max 64 GBApple · laptop	$4.1k	64 / 54 GB	614 GB/s	76 t/s	95 W	$332
MSI EdgeXpert MS-C931 (128 GB)MSI · small desktop	$4.7k	128 / 119 GB	273 GB/s	80 t/s	240 W	$887
NVIDIA DGX Spark (128 GB)NVIDIA · small desktop	$4.7k	128 / 119 GB	273 GB/s	80 t/s	240 W	$887
Single RTX A6000 48 GB (Ampere) buildNVIDIA · workstation	$4.7k	48 / 46 GB	768 GB/s	67 t/s	300 W	$1.0k
Single RTX 5090 buildNVIDIA · desktop tower	$4.9k	32 / 31 GB	1792 GB/s	165 t/s	520 W	$1.8k
Single AMD Radeon Pro W7900 48 GB buildAMD · workstation	$5k	48 / 46 GB	864 GB/s	62 t/s	295 W	$1.0k
Lenovo ThinkStation PGX (128 GB)Lenovo · small desktop	$5k	128 / 119 GB	273 GB/s	80 t/s	160 W	$650
Mac Studio M3 Ultra 96 GBApple · small desktop	$5k	96 / 80 GB	819 GB/s	98 t/s	180 W	$624
MacBook Pro M5 Max 128 GBApple · laptop	$5k	128 / 108 GB	614 GB/s	76 t/s	95 W	$332
2× Strix Halo cluster (256 GB unified)AMD · mini-PC pair	$6k	256 / 192 GB	256 GB/s	62 t/s	240 W	$841
Quad Intel Arc Pro B70 buildIntel · rack/large tower	$6k	128 / 124 GB	608 GB/s	89 t/s	700 W	$2.5k
Quad RTX 3090 (used) buildNVIDIA · rack/large tower	$6k	96 / 92 GB	936 GB/s	89 t/s	1400 W	$4.9k
Single RTX Pro 5000 Blackwell 48 GB buildNVIDIA · workstation	$7k	48 / 46 GB	1344 GB/s	—	300 W	$1.1k
Single RTX 6000 Ada 48 GB buildNVIDIA · workstation	$8k	48 / 46 GB	960 GB/s	98 t/s	300 W	$1.0k
Mac Studio M3 Ultra 256 GBApple · small desktop	$8k	256 / 232 GB	819 GB/s	98 t/s	180 W	$624
Dual AMD Radeon Pro W7900 buildAMD · workstation	$9k	96 / 92 GB	864 GB/s	80 t/s	600 W	$2.1k
2× DGX Spark cluster (256 GB unified, CUDA)NVIDIA · two desktops, 200 G interconnect	$10k	256 / 240 GB	273 GB/s	124 t/s	460 W	$1.7k
Dual RTX 5090 buildNVIDIA · rack/large tower	$10k	64 / 62 GB	1792 GB/s	178 t/s	1050 W	$3.6k
Octuple Intel Arc Pro B70 clusterIntel · rack/large tower	$11k	256 / 248 GB	608 GB/s	98 t/s	1450 W	$5k
4× Strix Halo cluster (512 GB unified)AMD · rack of 4 mini-PCs, 10 GbE fabric	$12k	512 / 384 GB	256 GB/s	84 t/s	480 W	$1.7k
Single RTX Pro 6000 Blackwell 96 GB buildNVIDIA · workstation	$12k	96 / 93 GB	1792 GB/s	178 t/s	600 W	$2.1k
Tinybox Red (6× 7900 XTX, 144 GB)tinycorp · 12U pedestal	$15k	144 / 138 GB	960 GB/s	142 t/s	1500 W	$5k
4× DGX Spark cluster (512 GB unified, CUDA)NVIDIA · rack of 4 desktops	$20k	512 / 488 GB	273 GB/s	133 t/s	920 W	$3.4k
8× Strix Halo cluster (1024 GB unified)AMD · rack of 8 mini-PCs, 10/25 GbE fabric	$23k	1024 / 768 GB	256 GB/s	111 t/s	960 W	$3.4k
Dual RTX Pro 6000 Blackwell buildNVIDIA · workstation	$24k	192 / 188 GB	1792 GB/s	196 t/s	1100 W	$3.8k
Single AMD Instinct MI325X 256 GB workstationAMD · workstation / 4U server (OAM)	$25k	256 / 250 GB	6000 GB/s	298 t/s	1000 W	$3.6k
Tinybox Green (6× RTX 4090, 144 GB)tinycorp · 12U pedestal	$25k	144 / 138 GB	1008 GB/s	196 t/s	2200 W	$8k
2× Mac Studio M3 Ultra 512 GB cluster (TB5 / MLX)Apple · two desktops, Thunderbolt 5 RDMA	$28k	1024 / 960 GB	819 GB/s	129 t/s	440 W	$1.5k
Single AMD Instinct MI300X 192 GB workstationAMD · workstation	$30k	192 / 188 GB	5300 GB/s	231 t/s	750 W	$2.8k
Single H100 80 GB workstationNVIDIA · workstation	$32k	80 / 78 GB	3350 GB/s	213 t/s	700 W	$2.5k
Quad RTX Pro 6000 Blackwell build (384 GB)NVIDIA · workstation / 4U pedestal	$38k	384 / 372 GB	1792 GB/s	284 t/s	2200 W	$8k
Single H200 141 GB workstationNVIDIA · workstation / 2U server	$40k	141 / 138 GB	4800 GB/s	298 t/s	700 W	$2.5k
Tinybox Pro (8× RTX 4090, 192 GB)tinycorp · 12U pedestal	$40k	192 / 184 GB	1008 GB/s	222 t/s	3200 W	$11k
8× DGX Spark cluster (1024 GB unified, CUDA)NVIDIA · rack of 8 desktops, 200 GbE fabric	$44k	1024 / 976 GB	273 GB/s	173 t/s	1840 W	$7k
Tinybox Green v2 (4× RTX 5090, 128 GB)tinycorp · 12U pedestal	$45k	128 / 124 GB	1792 GB/s	231 t/s	2300 W	$8k
8× RTX Pro 6000 Blackwell server (768 GB)NVIDIA · 4U server (e.g. SuperMicro AS-4125GS-TNRT)	$78k	768 / 744 GB	1792 GB/s	373 t/s	4800 W	$16k
8× H100 80 GB serverNVIDIA · server rack	$280k	640 / 620 GB	3350 GB/s	356 t/s	5600 W	$20k
NVIDIA RTX Spark (128 GB)NVIDIA · OEM laptops + small desktops	—	128 / 119 GB	300 GB/s	—	— W	$756

Last updated 2026-07-08