Of course, they don have sharding now, as they still dont have maiinet. They will launched maiinet on December 14th this year.
They won't have sharding in their main net. They state they are researching about sharding. They didn't implement it. Please check the facts. They have "state sharding" in their roadmap. For your information, state sharding is what ethereum aims for but it is theoretical, still open to debate, it is not even known if it is possible to implement it or not. So they have nothing in hand about sharding, they won't have it in main-net launch.
Even though they somehow achieved implementing sharding for their v2 launch, it is months later than Zilliqa's main-net. And as I said, there is no indication that they will able to implement it.
Zilliqa mainnet will be launched with the sharding (network sharding). Zilliqa's sharding blended in their project from the beginning.
I don't dismiss their team, I shouldn't since I don't know them. I'm basing my opinion according to their GitHub repo. So, there is nothing new, unique, or different about Tomochain regards to their mainnet launch. I don't know how it will evolve later. But currently, they are not a competitor of Zilliqa. Zilliqa is in a different league. Zilliqa's team amazing. I mean, for example, Ilya Sergey regarded very highly by academics in his field. I won't compare the teams but if you have a good understanding of programming, just check their repos and there won't be anything to discuss. Zilliqa is superior in every way possible. Again, I'm talking about the current state of the developments.
P.S. Btw, Zilliqa has already over 10k TPS in their stress tests when limited to the payment transactions. And this is in a decentralized manner. Not like having some master nodes to process transactions i.e. similar to distributed SQL (TomoChain, EOS, NEO). So I wasn't focused on TPS since Zilliqa is way above than others if you stick with decentralization target. And plenty of room to improve. No other chain comes close to them as far as I know.