Piotr Lewandowski

> Is the new output compatible with serialization produced with builtin json? Almost always yes. `read` methods were not touched, and the tests are passing. The serialization output is not...

I'd like to work on this one

As I first step, I will benchmark [`orjson`](https://github.com/ijl/orjson) in as simple as possible way to see if we can consider staying within JSON model. `orjson` is considered most performant (and...

After very initial investigation, `orjson` seems like agood place to start. My goal is to have < 150 ms for 1000 qubit, 100 moment performance check. I'll prepare draft PR...

**Important note is that `orjson` serialize to bytes and not string** A quick result from very simple introduction of `orjson` can be seen below. From profiling (picture below) we can...

@rht, @tanujkhattar - as stated before I prepared draft PR - https://github.com/quantumlib/Cirq/pull/6115. Copying the initial description for your convince 1. After ensuring that I changed as little as possible it...

@rht, those are very valid points. I observed another thing that might be worth taking into consideration. If we omit [`ContextualSerialization`](https://github.com/quantumlib/Cirq/blob/ebec38b807f4463f803df136472c2ad1ffb4f369/cirq-core/cirq/protocols/json_serialization.py#L613-L636) (I simply commented it out) then we get massive...

https://github.com/quantumlib/Cirq/pull/6115 is ready to be reviewed. I believe resulting performance meets the requirements. Cc: @rht @tanujkhattar

Given that the PR is already prepared, I'd like to highlight the fact that there is limited downside for migrating to `orjson`. The downsides are (the one that I am...

Hello, I'd like to take a stab at this issue. I was not involved before. Will read through current implementation and comment history and then provide a comment with my...