# Multiple VTA instances

**URL:** <https://discuss.tvm.apache.org/t/multiple-vta-instances/17678>\
**Category:** Questions\
**Created:** [September 3, 2024, 1:59am UTC](https://discuss.tvm.apache.org/t/multiple-vta-instances/17678 "2024-09-03T01:59:13Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![leokonto](https://discuss.tvm.apache.org/letter_avatar_proxy/v4/letter/l/4af34b/32.png) [@leokonto](https://discuss.tvm.apache.org/u/leokonto)\
**Post date:** [September 3, 2024, 1:59am UTC](https://discuss.tvm.apache.org/t/multiple-vta-instances/17678/1 "2024-09-03T01:59:13Z")

</div>

Hi, I wanted to ask if it is possible to create multiple instances of the VTA (fast or cycle accurate) simulator and/or on a target FPGA. I would like to use the `pipeline executor` to concurrently run data independent graphs. I have seen this previous post ([Can I deploy multiple VTAs and run those on single PYNQ board?](https://discuss.tvm.apache.org/t/can-i-deploy-multiple-vtas-and-run-those-on-single-pynq-board/5091)), but I am using the PolarFire FPGA, which could resource-wise support multiple instances. Drivers for the PolarFire FPGA have also been implemented to support execution.

Thanks in advance.