Qwen3.5 397B A17B — Hardware Requirements & GPU Compatibility
VisionFunctions
Qwen3.5 397B A17B is a 403.4B-parameter open language model from Alibaba in the Qwen 3.5 family. It supports a context window of up to 262,144 tokens. At Q4_K_M it needs about 242.46 GB of VRAM — see which GPUs and Macs can run it below.
Qwen3.5 397B A17B runs on a 256 GB+ Mac or a datacenter GPU
Best local pick → Mac Studio (M3 Ultra, 256GB) · Q4_K_M · 242.6 GB · ~27 tok/s
Macs from 128 GB (IQ2_XS · ~40 tok/s)
Score 113#20 of 100 open#9 of 84 that fit ≤256 GBApache 2.0
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
llmrun Composite Score →
113.0
- 90% range 109.0–117.0
- #20 of 100 open models
- #9 of 84 that fit ≤256 GB
Run Qwen3.5 397B A17B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
ollama run hf.co/unsloth/Qwen3.5-397B-A17B-GGUF:Q4_K_MQ4_K_M runs on one Instinct MI325X 256 GB · 16K context
Download: Q4_K_M/Qwen3.5-397B-A17B-Q4_K_M-00001-of-00006.gguf244.09 GB + 5 partsunsloth
Other ways
each opens step-by-stepRun in cloud
Fits on B300 (288 GB) (45 GB headroom) · Q4_K_M
- Generation speed
- ~163 tok/s
- generation speed
- Cost per 1M output tokens
- $11.80
- per 1M output tokens
How Much VRAM Does Qwen3.5 397B A17B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 171.9 GB | 172.3 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Runs on B200 192 GB | |
| 176.9 GB | 177.3 GB | 3-bit small — 3-bit small quantization | Runs on B200 192 GB | |
| 197.1 GB | 197.5 GB | 3-bit medium — 3-bit medium quantization | Runs on Instinct MI325X 256 GB | |
| 202.1 GB | 202.5 GB | 4-bit · legacy — 4-bit legacy quantization | Runs on Instinct MI325X 256 GB | |
| 242.5 GB | 242.8 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Runs on Instinct MI325X 256 GB | |
| 287.9 GB | 288.2 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Runs on Instinct MI350X 288 GB | |
| 333.2 GB | 333.6 GB | 6-bit, very good — 6-bit quantization, very good quality | Runs on 5× H100 SXM 80 GB | |
| 403.8 GB | 404.2 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Runs on 6× H100 SXM 80 GB |
Selected Q4_K_M
4-bit medium quantization242.6GBat 16K context
- Weights
- 242.0 GB
- KV cache
- 0.3 GB
- Overhead
- 0.3 GB
Server-class: ~242.6 GB at 16K context · one Instinct MI325X 256 GB
Which Hardware Can Run Qwen3.5 397B A17B?
Q4_K_M · 242.5 GBQwen3.5 397B A17B (Q4_K_M) requires 242.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 316+ GB is recommended. Using the full 262K context window can add up to 3.9 GB, bringing total usage to 246.4 GB. No consumer GPU has enough memory. 2 devices with unified memory can run Qwen3.5 397B A17B, including Mac Studio (M3 Ultra, 512GB), Mac Studio (M3 Ultra, 256GB).
128–144 GB
Q4_K_M · best fit IQ2_S (126.6 GB)
4 GPUs
GH200 Grace Hopper Superchip~175Simulate output on GH200 Grace Hopper SuperchipH200 NVL~174Simulate output on H200 NVLH200 SXM~174Simulate output on H200 SXM192+ GB
Q4_K_M · best fit IQ3_M (182.1 GB)
6 GPUs · 1 Mac class · 2 device families
B200~185Simulate output on B200B300~185Simulate output on B300Instinct MI350X~70Simulate output on Instinct MI350XM3 Ultra256–512 GB~32Simulate output on Mac Studio (M3 Ultra, 256GB)8x H100 SXM5~254Simulate output on NVIDIA DGX H1008x A100 SXM4~231Simulate output on NVIDIA DGX A100 640GB
By VRAM tier · typical GPUs and Macs at Q4_K_M
Consumer GPUs (up to 32 GB) and Macs up to 96 GB: too small for this model (needs ≥ 111.5 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
How good is Qwen3.5 397B A17B?
The llmrun Score merges 8 public benchmark boards into a single figure, where 100 represents the average for open models.
Score vs VRAM in the ≤256 GB tier
- Qwen3.5 397B A17B
- Other models
- Best Score for its VRAM
#9 of 84 that fit ≤256 GB
Each dot is a model at the best quant that fits ≤256 GB. The line joins the models with the best Score for their VRAM; a dot below it is beaten by a model that needs no more memory.
Full ≤256 GB ranking →Benchmarks
The 6 boards that weigh most in the llmrun Score, of 10 with a result for this model; badges show who ran the test.
- GPQA Diamond
Epoch AI · Feb 2026Independent
Independent86.4%
#18/98 open
- LMArena Text
LMArena · Oct 2026Arena
Arena1441.6
#17/187 open
- AIME 2024/2025
Epoch AI · Feb 2026Independent
Independent88.9%
#16/86 open
- FrontierMath
Epoch AI · Feb 2026Independent
Independent31.2%
#16/17 open
- LMArena WebDev
LMArena · Oct 2026Arena
Arena1400.5
#28/52 open
- MMLU-Pro
TIGER-LabIndependent
Independent87.8%
#2/145 open
In the ≤256 GB class
84 models fit ≤256 GBSame Score, only models that fit ≤256 GB at their best quant: the ranking a ≤256 GB owner actually chooses from.
- #1
MiMo V2.6 Flash RLXiaomi
131.9
~83 tok/s · 186.5 GB
- #2
DeepSeek V4.1 FlashDeepSeek
129.5
~45 tok/s · 210.4 GB
- #3
DeepSeek V4 Flash 0731DeepSeek
127.6
~76 tok/s · 183.0 GB
- #4
GLM 5.3 FlashZ.ai
124.2
~83 tok/s · 205.2 GB
- #5
GLM 5.2Z.ai
122.9
~52 tok/s · 227.8 GB
- #6
Qwen3.8 27BAlibaba
122.1
~207 tok/s · 17.9 GB
Alternatives to Qwen3.5 397B A17B
Qwen 3.5 family · 34 models →Official models with an llmrun Score, each answering one question; sizes are each model's default quant (usually Q4_K_M) with a 16K-token context.
Best alternative at this size
MiMo V2.6 Flash RL
Xiaomi
- 18.9 higher Score
- 56.1 GB less VRAM
Score 131.9 · 186.5 GB
Fits in less VRAM
Inkling Small
Thinking Machines
- 6.6 higher Score
- 79.9 GB less VRAM
Score 119.6 · 162.7 GB
Other vendor
GLM 5.3
Z.ai
- 17.1 higher Score
- 211.2 GB more VRAM
Score 130.1 · 453.8 GB
Similar models
10| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| DeepSeek V4.1 FlashDeepSeek | 129.5 | 458.4 GB | —needs > 24 GB | similar score |
| DeepSeek V4 Flash 0731DeepSeek | 127.6 | 183 GB | —needs > 24 GB | similar score |
| GLM 5.3 FlashZ.ai | 124.2 | 205.2 GB | —needs > 24 GB | similar score |
| GLM 5.2Z.ai | 122.9 | 453.8 GB | —needs > 24 GB | similar score |
| GLM 5.1Z.ai | 117.8 | 454.1 GB | —needs > 24 GB | similar score |
| MiniMax M3MiniMax | 112.5 | 258 GB | —needs > 24 GB | similar score |
| MiniMax M2.5MiniMax | 111.8 | 139.6 GB | —needs > 24 GB | similar score |
| GLM 5Z.ai | 111.6 | 454.1 GB | —needs > 24 GB | similar score |
| Qwen3.6 27BAlibaba | 111.2 | 17.9 GB | ~38 | similar score |
| DeepSeek V3.2DeepSeek | 110.4 | 412.7 GB | —needs > 24 GB | similar score |
Fine-tunes & variants(1)
Specifications
- Publisher
- Alibaba
- Family
- Qwen 3.5
- Parameters
- 403.4B
- Active parameters
- 24.4B per token
- Architecture
- Qwen3_5MoeForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 248,320
- Release Date
- 2026-02-16
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Qwen3.5 397B A17B need?
Qwen3.5 397B A17B requires 242.5 GB of VRAM at Q4_K_M, or 807.2 GB at BF16. Full 262K context adds up to 3.9 GB (246.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 403.4B × 4.8 bits ÷ 8 = 242 GB
KV Cache + Overhead ≈ 0.5 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 ≈ 4.4 GB (at full 262K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run Qwen3.5 397B A17B?
No — Qwen3.5 397B A17B requires at least 111.4 GB at IQ2_XXS, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
What's the best quantization for Qwen3.5 397B A17B?
For Qwen3.5 397B A17B, Q4_K_M (242.5 GB) offers the best balance of quality and VRAM usage. Q4_K_L (247.5 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 111.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Qwen3.5 397B A17B on a Mac?
Yes — Mac Pro M2 Ultra (128 GB) and 14 other Macs can run Qwen3.5 397B A17B, the smallest at IQ2_XS. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 242.5 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio (M3 Ultra, 256GB).
Can I run Qwen3.5 397B A17B locally?
Yes — Qwen3.5 397B A17B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Qwen3.5 397B A17B?
At Q4_K_M, Qwen3.5 397B A17B can reach ~66 tok/s on AMD Instinct MI350X. 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 B300 → 14.7 GB active ÷ (8000 × 0.65) = 2.82 ms, plus 60 layers × 0.055 ms = 3.30 ms, so 1000 ÷ 6.12 ms = ~163 tok/s
Estimated speed at Q4_K_M (242.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Qwen3.5 397B A17B?
At Q4_K_M, the download is about 242.04 GB. The full-precision BF16 version is 806.80 GB. The smallest option (IQ2_XXS) is 110.93 GB.
Which GPUs can run Qwen3.5 397B A17B?
No single consumer GPU has enough VRAM to run Qwen3.5 397B A17B at Q4_K_M (242.5 GB). Multi-GPU or professional hardware is required.
Which devices can run Qwen3.5 397B A17B?
2 devices with unified memory can run Qwen3.5 397B A17B at Q4_K_M (242.5 GB), including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.