If there's anything I've learned building extraction agents is that proxies are pretty... sketch.
And on that vein, I'm super stoked to disconnect my smart tv, ditch the sluggish OS, and get an AppleTV.
The latest Rex preview release includes a very important subsystem I've been working on for weeks to enable full forwards+backwards API compatibility with clients and servers. This was gating the Linux test program and massively opening up the Mac test program.
We just invited another 1,000 (to Mac) and if it goes well this will step up considerably this week. The Linux test program will likely start this week (separate invite waves amongst existing testers).
The API compatibility system is super cool. At a high level:
Every API change is a rule on a JSON pointer (RFC 6901).
For each message, we compile all the rules between your version and ours into a trie keyed by JSON path tokens. This happens lazily since we don't want a full quadratic explosion (a message with N revisions has N possible chains of length 1 to N). We don't want to build N^2/2 trie nodes when most WON'T be visited.
Next, we wrote a custom recursive descent streaming JSON editor (forward-pass only). If a key doesn't match any subtree in the trie, we fast-path copy the full value onto the wire (nothing in it can possibly matter for compatibility).
Equal chains of changes are interned to the same pointer, so when we broadcast an event to many old clients, we translate it once per distinct chain, not once per client.
The result of all this plus some other work is that we simulated various shapes of 10,000 long revision chains (note: we're on revision 4 at the moment...) with pathological changes of various shapes and an upgrade costs about 35 _microseconds_ on a 20KB JSON body.
Opus 5.5 fast mode in the Claude Platform is no longer in research preview, and weβve removed the waitlist!
Itβs the same Opus 5.5 but up to 2.5x faster, at $8 / $40 per million tokens.
It's also available in Microsoft Foundry (Hosted on Azure), with Google Cloud coming soon.
Fast mode in Claude Code remains unchanged, and uses extra usage credits.
See our docs here:
https://t.co/RikPqqIY0W
https://t.co/oB0fhEgC88
Anyone else seeing Sol-6.1 from @OpenAI hallucinating token budgets within reasoning traces?
Using Sol through the API, current harness does not expose tokens, limits, etc