Qwen3.5 9B

Unsloth's Q4_K_M is 5.7 GB and Q8_0 is 9.5 GB. Unsloth's sizing table puts 4-bit at about 6.5 GB of total memory once runtime overhead is included.

Running it

8 GB GPU at Q4_K_M, a 12 GB GPU at Q8_0, or any 16 GB Mac. See the hardware sizing tables for how that maps to specific chips and cards, and the quantisation guide for what you give up at each bit width.

Which cloud instance actually runs this, and what it costs

At Q4_K_M the weights are about 5.7 GB, so 20 catalogued instances fit with room for context, of which the 8 best value are shown. Cheapest per token is Hetzner GEX45 (~$0.68 per million output tokens at an estimated 118 tok/s). Fastest on a single card is Azure NCads H100 v5 at roughly 588 tok/s for $3.30 per million tokens, which is the speed-versus-cost trade in one line.

Token generation on a single request is memory-bandwidth-bound, not compute-bound: the GPU must read every active weight from memory for each token it emits. So the ceiling is roughly GPU memory bandwidth divided by the size of the active weights. A 16 GB model on a 300 GB/s L4 tops out near 19 tokens/sec; the same model on a 2039 GB/s A100 tops out near 127. Figures on model pages are that arithmetic, computed from the bandwidth and weight-size columns here. They are a ceiling for one request at batch size 1, before serving overhead, so treat them as an upper bound for comparing instances rather than a benchmark. Batching raises total throughput well above this and does not raise per-request speed.

For identical silicon, the hyperscalers charge roughly 2 to 5 times what the specialist GPU clouds charge. An A100 80GB is about $3.43/GPU-hr on AWS, $3.67 on Azure and $5.03 on Google Cloud, against $1.19-1.39 on RunPod and $1.99 on Lambda. That spread is the single largest cost lever in a self-hosted inference budget, and it is larger than any saving from picking a smaller model.

Three costs that are absent from every headline rate: egress, storage for the weights, and idle time. A 70B model at Q4 is tens of gigabytes that must be pulled to the instance on every cold start, and an endpoint billed hourly is billed while it sits idle waiting for a request. Rates read 2026-09-06; check each provider's live pricing on the directory hub.

Jurisdiction: China

This is the crucial split most write-ups miss: using a Chinese lab's HOSTED API sends your data to Chinese infrastructure and is a real procurement question. Downloading their OPEN WEIGHTS and running them on your own hardware, or on a Western host, sends nothing anywhere. DeepSeek and Z.ai publish under MIT; Alibaba publishes most of Qwen3 under Apache 2.0. Those are among the most permissive licences on this page.

Watch for: Several US states and a number of government bodies restrict Chinese-hosted AI services on official devices. That restriction is about the hosted service, not about the weights running in your own VPC — but expect to have to explain the difference to a security reviewer.

Frequently asked

Similar models