A16
The A16 is a cloud GPU with 16 GB of memory. It is currently available from 1 provider starting at $1.02/hr, with a market median of $1.02/hr across 1 offering.
Hardware specifications
- Architecture
- Ampere
- VRAM
- 16 GB
Same across all providers.
Market comparison
Offerings
No offerings found
Try clearing the spot filter or check back later.
LLMs that fit
| Model | Params | 16-bit | 8-bit | 4-bit |
|---|---|---|---|---|
Laguna-S 2.1 Poolside | 118B 8B active | Needs >8 GPUs | Needs >8 GPUs | Barely fits on 7× |
gpt-oss-120b OpenAI | 117B 5.1B active | Not released | Not released | Barely fits on 6× |
Llama 3.1 70B Meta | 70B | Needs >8 GPUs | Barely fits on 6× | 4× · 10 GB free |
Qwen3.6 35B-A3B Alibaba | 35B 3B active | Barely fits on 7× | Barely fits on 4× | 3× · 14.4 GB free |
Qwen3 32B Alibaba | 32.8B | Barely fits on 6× | 4× · 12.4 GB free | Barely fits on 2× |
Qwen2.5 14B Alibaba | 14B | 3× · 12.6 GB free | 2× · 12.7 GB free | 1× · 4.6 GB free |
Llama 3.1 8B Meta | 8B | 2× · 11.1 GB free | 1× · 4.6 GB free | 1× · 9.3 GB free |
Mistral 7B Mistral AI | 7B | 2× · 13.8 GB free | 1× · 6 GB free | 1× · 10.1 GB free |
Qwen3 4B Alibaba | 4B | 1× · 4.4 GB free | 1× · 9.6 GB free | 1× · 12.3 GB free |
Estimates: model weights plus 20% for KV cache and runtime overhead, at short context lengths. Long contexts and large batches need more. Rows marked “tight” hold the weights but not that full margin. A ×N figure is the total VRAM across N of these GPUs and makes no claim about interconnect throughput.
Similar GPUs
See more GPUs like this in
Frequently asked questions
How much does the A16 cost per hour?
The cheapest A16 offering starts at $1.02/hr, and the market median is $1.02/hr.
Which cloud providers offer the A16?
No providers currently list the A16 in our catalog.
How much VRAM does the A16 have?
The A16 has 16 GB of VRAM.
What LLMs can run on the A16?
With 16 GB of VRAM, the A16 can run 9 of our tracked models in 4-bit quantization, including Laguna-S 2.1, gpt-oss-120b, Qwen3.6 35B-A3B. 6 models fit in 16-bit precision.
What's the biggest LLM I can run on multiple A16s?
With 8× A16 (128 GB total), the largest tracked model that fits in 16-bit is Qwen2.5 14B (14B params), which needs 3× GPUs.