Skip to content

Qwen3.8 Flash-Next on M3 Ultra 96 GB: VRAM, speed and fit

Qwen3.8 Flash-Next at 2-bit runs on the M3 Ultra 96 GB with 192K of context at about 90 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on M3 Ultra 96 GB · 2-bit (Q2_K) · Ollama / LM Studio · Chat

GPU memory
92.3 GB
of 96.0 GB
Context
192K
auto
Speed
90 tok/s
49 when full
First reply
3.2 s
2K prompt

Where the memory goes

Model
62.9 GB
Context
4.6 GB
Overhead
821 MB
OS
24.0 GB
Free
3.7 GB
Memory
92.3 GB of 96.0 GB
Context
192K tokens · auto
Speed
90 tok/s · 49 when full
First reply
3.2 s for a 2K prompt
Running cost
$3.55/mo at home · $122/mo on a cloud 4× RTX A4500

Validation: Validated with conditions

What Qwen3.8 Flash-Next on M3 Ultra 96 GB was checked against
CheckTargetResultStatus
Fits in GPU memory72.0 GB usable68.3 GB neededPass
Headroom5% free3.7 GB free (5%)Pass
Context for chat16K or more192KPass
Writing speed15 tok/s or more90 tok/sPass
First replyunder 3 s, 2K prompt3.2 sCheck

Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.

The setup

Model
Qwen3.8 Flash-Next
Parameters
180B, 6B active
Precision
2-bit (Q2_K)
Context
192K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
M3 Ultra 96 GB
GPU can use
72.0 GB of 96.0 GB

Running cost

At home
$3.55/mo
162 W at the wall
Renting instead
$122/mo
4× RTX A4500 at $1.00/h

4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 22, 2026.

How to run it

Ollama
ollama run qwen3.8-flash-next:125b-a6b-q4_K_M
llama.cpp
llama-server -hf unsloth/Qwen3.8-Flash-Next-GGUF
vLLM
vllm serve Qwen/Qwen3.8-Flash-Next
MLX (Mac)
mlx_lm.chat --model mlx-community/Qwen3.8-Flash-Next-4bit

Qwen3.8 Flash-Next on similar machines

All machines: Qwen3.8 Flash-Next VRAM requirements. Everything that runs here: what can the M3 Ultra 96 GB run?

Other models that run on M3 Ultra 96 GB

Questions

Can Qwen3.8 Flash-Next run on M3 Ultra 96 GB?

Qwen3.8 Flash-Next at 2-bit runs on the M3 Ultra 96 GB with 192K of context at about 90 tokens a second.

Report VR-0UZF3PZ, calculated Nov 22, 2026 from published specs, not benchmarked.