The only AI gateway with built-in observability and evals. respan.ai/discord
San Francisco, California
- Tweets 719
- Following 75
- Followers 2,774
- Likes 1,366
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
115
126
1,804
278,798
Can Jev answer 83,719 × 6,283? No.
Span-01 can.
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
1
1
13
363
We’re rolling out Span-01 access now. Keep an eye on your inbox!
Many of you asked about latency. Span-01 is smaller than Jev and much faster!
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
14
2
81
3,699
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
115
126
1,804
278,798
come hang out in our discord 👋
we’d love to hear how you’re using Span-01, what’s working, and what we can make better.
respan.ai/discord
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
7
2
21
1,485
Respan retweeted
ok i wasn't expecting this many of you on the Span-01 waitlist 🥹
can't wait to get it into your hands.
while you wait… we wrote it a song. yes, our model has a song now.
sound on!
Introducing Span-01, the first hyper-parallel reasoning classifier built for unseen challenges (RLAIF).
2x cheaper, 18% better than Jev.
700x cheaper, 4% better than GPT-6 Luna.
Frontier reasoning for every behavior, at classifier speed.
• Span-01: #1 on Behavior Benchmark
• Span-01 Lite: Better than Jev and completely free!
8
10
27
4,385
A user reports a bad answer. You have the output, but no idea which step caused it.
Respan traces every run so you find the step that broke, then watches cost, errors, and user behavior across all of them so you catch the next one early.
Here's the full walkthrough of Respan Observability 🧵
5
1
13
863
More walkthroughs like this on our YouTube channel:
piped.video/@RespanAI
1
60
At @RespanAI, we care about security as much as performance, @francchen put together a demo of how prompt injection can change Jev’s answer.
We’re building something better. 🔨
I don’t know if people still remember Jev. Things move so fast here.
I’ve spent the last few days testing it, and found a few things I think builders should see.
Prompt injection is one of the most interesting things to test in AI, so I made a little demo to show how it could change Jev’s answer.
Jev is a great model. I just want people to know what to watch out for when they use it.
1
12
922
