HereBePandas

Yes, two things: 1. GPT-5.1 Codex is a fine tune, not the "vanilla" 5.1 2. More importantly, GPT 5.1 Codex achieves its performance when used with a specific tool (Codex CLI) that is optimized for GPT 5.1 Codex. But when labs evaluate the models, they have to use a standard tool to make the comparisons apples-to-apples.

Will be interesting to see what Google releases that's coding-specific to follow Gemini 3.

You're right on SWE Bench Verified, I missed that and I'll delete my comment.

GPT 5.1 Codex beats Gemini 3 on Terminal Bench specifically on Codex CLI, but that's apples-to-oranges (hard to tell how much of that is a Codex-specific harness vs model). Look forward to seeing the apples-to-apples numbers soon, but I wouldn't be surprised if Gemini 3 wins given how close it comes in these benchmarks.

Not apples-to-apples. "Codex CLI (GPT-5.1-Codex)", which the site refers to, adds a specific agentic harness, whereas the Gemini 3 Pro seems to be on a standard eval harness.

It would be interesting to see the apples-to-apples figure, i.e. with Google's best harness alongside Codex CLI.

[comment removed]

They explicitly address this in page 11 of the report. Basically perfect recall for up to 1M tokens; way better than GPT-4.

Yes

Weird - just tried this and it worked for me.

Tech report seems to hint at the fact that GPT-4 may have had some training/testing data contamination and so GPT-4 performance may be overstated.

I'd be shocked - given the incentives - if it hasn't already happened to a great extent. Many of the types of people Google DeepMind hires are also the types of people hedge funds hire.

Sure, but this time in 2022, you probably wouldn't have let it do that much (or even had this thought).

Progress!

The US has had the Bill of Rights and its constitutional system / property protection for a long time but only recently has had the degree of NIMBYism that it's had.

It's easy to pretend the bug is actually an important and noble feature, but sometimes it's just a bug that needs fixing.