rohansood15

But you need to explicitly specify the prompt to check dosage. If you passed it as part of the original extraction model call, it would return a lower confidence to begin with. Or provide a different 'is_dosage_reasonable' value, which would eliminate the need for Jev.

And this is before one argues that dose <= 3 tablets is a better validation than the Jev prompt and costs nothing.

Jev is cool, but this example is really bad.

The example flags a '10 tablets' dose as a possible erroneous extraction, when the text says exactly that. So it is a false positive.

Furthermore, even if your goal was to catch logical errors rather than extraction errors, you are better off using programmatic checks on the extracted value and having a dedicated error handling path including additional full LLM calls, rather than decision model.

Can you share an actual example of where it works please?

Wishing you guys the best, but this isn't a very compelling answer. Would be helpful to provide more concrete or quantifiable examples. In the early stages when pitching to users, it's preferable to state exactly what can be done now.

Because labs can learn to optimize inference post launch, plus can move to use bigger/better clusters depending on demand. It is not impossible to imagine Qwen cuts prices further with QAT/MTP-like improvements.

Given the new architecture, speeds are harder to estimate. Max would be ~40 tok/s.

This is a good counter argument. But you have to note that this is after OpenAI cut Luna costs by 80%. If you compare launch pricing, Qwen probably comes out ahead on a cost-performance basis.

128GB, 4-bit quantized.

For world knowledge, you'd want it to find and reference the source material to be sure. At that point, it doesn't matter if the knowledge is embedded.

Didn't expect it to beat 3.8 27B so cleanly.

Opus 4.6 Max self-hosted at 30 tok/s on a 5k Macbook in Aug 2026. The LLM timelines are crazy.

So an engineer who knows your codebase and tests can sneak in malicious code/backdoors because you're high trust.

And I am guessing you'll extend that high trust to your agents/'orbs' next. And I am sure you'll find an auditor who'll go with it coz frankly most don't care.

I am not surprised there are folks willing to do this, but I am surprised that you feel you must brag about it. And you ARE bragging when you title the post the way you did. Good luck.

Companies have learned their lessons on stickiness with cloud providers. Every enterprise has a multi-provider strategy now.

I agree with you, but Codex is open source.