I've been looking at Targon (SN4) and I'm curious how it stacks up against other inference providers on the network. Has anyone benchmarked latency or throughput? Especially interested in real-world API usage.
I've been benchmarking Targon vs Apex. Targon is faster for raw inference but Apex has better output quality. Depends on your use case.
From a business perspective, Targon's API is more mature. If you need programmatic access, it's the better choice right now.
Thanks Orion! That's exactly the comparison I was looking for. Will try both side by side.
