GMKtec EVO-X2 (128GB unified)
The $/GB winner: runs 70B-class and quantized big-MoE (V4 Flash tier) that no consumer GPU fits
A100 SXM 40GB cloud pricing, August 2026
Live A100 SXM 40GB quotes (40 GB VRAM) across the providers we track, cheapest first. Marketplace prices are real asks from the Vast.ai order book; secure/on-demand tiers carry datacenter SLAs (what the tiers mean).
Quotes refreshed 2026-08-23
Price floor: $0.388/hr (marketplace) at
Vast.ai.
Rent a A100 SXM 40GB →
| Provider | Tier | $/hr per GPU | Median $/hr | Live offers | |
|---|---|---|---|---|---|
| Vast.ai | marketplace | $0.388 | $0.800 | 18 | Rent |
| Verda | spot | $0.645 | - | - | |
| RunPod | community | $1.00 | - | - | Rent |
| Verda | on-demand | $1.29 | - | - |
What fits on a single A100 SXM 40GB
Models from our tracked open-weight set that fit in 40 GB, with KV-cache headroom included. Bigger models need multi-GPU nodes (math in the calculator).
| Model | Params | Fits at | VRAM needed | Cheapest API $/1M out |
|---|---|---|---|---|
| Qwen3.6 27B | 27.8B | FP8 | 39 GB | compare → |
| Qwen3.5 27B | 27.8B | FP8 | 39 GB | compare → |
| Mistral Small 3.1 | 24B | FP8 | 33 GB | |
| IBM Granite 4.1 8B | 8.8B | FP8 | 13 GB | compare → |
| GPT-OSS 20B | 21B | FP8 | 29 GB | compare → |
| Llama 3.1 8B | 8B | FP8 | 11 GB | compare → |
| Llama 3.2 3B | 3.2B | FP8 | 5 GB | compare → |
| Gemma 3 27B | 27B | FP8 | 38 GB | compare → |
| Gemma 3 12B | 12B | FP8 | 17 GB | compare → |
| Phi-4 | 14.7B | FP8 | 21 GB | compare → |
| GLM-4.7 Flash | 31B | INT4 | 22 GB | compare → |
| Qwen3.6 35B A3B | 36B | INT4 | 25 GB | compare → |
| Qwen3.5 35B A3B | 36B | INT4 | 25 GB | compare → |
| Qwen3 32B | 32.8B | INT4 | 23 GB | compare → |
| Qwen3 30B A3B | 30.5B | INT4 | 21 GB | compare → |
| Nemotron 3 Nano | 30B | INT4 | 21 GB | compare → |
| IBM Granite 4.1 30B | 30B | INT4 | 21 GB | |
| Mixtral 8x7B | 46.7B | INT4 | 33 GB | compare → |
| DeepSeek R1 Distill Qwen 32B | 32.8B | INT4 | 23 GB | compare → |
Can't buy a A100 SXM 40GB? The desk-side alternatives
A100 SXM 40GB cards only ship in servers, but the "own your inference box" itch has real answers: run the economics in the calculator, then look at what this audience actually buys:
NVIDIA RTX 5090 (32GB)
Bandwidth-per-dollar king for the A3B-MoE and 27-32B dense workhorses
NVIDIA RTX PRO 6000 Blackwell (96GB)
The startup standard: only sub-$10K single card that runs 70B+ comfortably
As an Amazon Associate we earn from qualifying purchases · hardware links may be referral links · picks are editorial, not paid · disclosure
FAQ
How much does it cost to rent a A100 SXM 40GB?
As of 2026-08-23, A100 SXM 40GB rentals start at $0.388/hr (marketplace tier at Vast.ai). Datacenter capacity with SLAs typically costs more than community or marketplace hardware.
What can you run on a A100 SXM 40GB?
With 40 GB of VRAM, a single A100 SXM 40GB fits models up to roughly 28B parameters at FP8 or 57B at INT4 quantization, including KV-cache headroom.
Cite or embed today's floor
Live badge for a README. It updates itself with every daily refresh:
[](https://llmhosting.ai/gpus/a100-sxm-40gb)
Citation: A100 SXM 40GB rental floor $0.388/hr (llmhosting.ai, 2026-08-23). Free to reuse with a link back.