I've just renamed my files from CLAUDE.md to AGENTS.md I certainly won't do that again regressing back in the vendor specific direction lol

This would make sense. In our local coding agent battle system, Argon is tanking, and this is running on opencode, importing agents.md and claude.md.

I find their products confusing as usual. Does Antigravity sub work same as $20 Codex? Are the limits somewhat similar to ChatGPT and Claude?

I asked Gemini to review the repo. Apparent issues include:

  - Corrupted/dropped Gemini thought_signature (reasoning_details) in tool loops
  - Blinded model.api.id = "contestant" bypasses gemini.txt and serves Claude's prompt (default.txt)
  - Anthropic-only prompt caching

Prompt was:

> I read on hacker news the following, which does not match my expectations. Check if something in the harness or bench is hurting gemini perf.

followed by your post.

I've used argon via `agy` cli tool and was not impressed at all. However, I decided to burn up the remainder of my ultra credits on a visual reasoning pet project.

Argon in low and medium mode had higher quality results than it did when running in high mode, almost every time. My tasks involved the agent finding some details in a raster image and then mark it up with predictions about what it found.

Argon on high would do the work, and then a whole bunch of other stuff i did not ask for. All that extra broke everything else downstream from the analysis.

However, doing the same work with Argon in low effort was on par with results from Haiku 5.5 (xhigh) and GPT-6.1-Terra (medium). So, Argon low works better than Argon high, but for my tasks -- a very weak model relative to how google positions it.

I'm ready for Geimini Unobtanium!

That doesn't seem to match the arena.ai rankings, who I'm assuming has a far larger dataset of rankings than you currently hold.

https://arena.ai/leaderboard

Well, my perspective is somewhat narrow (I use it primarily for software engineering and pretty technical data engineering work), so it's possible that Argon is well-suited for the work I do but not for other tasks. I suppose it's just more data points to consider.

This does not mesh with other battles like Arena where Argon is nearer to the top.

Have you considered that the harness might be the issue?

> local coding agents

what does "local" coding agents mean? Surely you're not hosting it..

my anecdotal experience is that argon and muse wonder full are both a full tier below opus/sol.

Haven’t tried opus 5, sol 6 or haiku 4.5 I see

Because Google’s models have been a disappointment for so long (Argon tentatively excluded) that nobody is excited to read a press release from them stating a new one is coming out “soon.” There’s no hype when your product to date (tentatively pre Argon) has been inferior for a “long time.”