Comment by manlymuppet
15 days ago
People should appreciate Google more for how all of their models have:
- Natively multi-modality
- Enormous context windows
- Reliably cheap prices
It's like they don't release a model until it passes the above checklist.
Most providers have some mixture of these three occasionally, but with Google, every model is consistent. This kind of consistency is really nice to have.
And the most reliable. A year and a half ago this was the exact reason why I switched to Gemini API from OpenAI. OAI would make me wait up to 10 minutes for the first token in the most egregious cases. Gemini served me almost immediately every time.
Huh? The constant failures in the Gemini API and overnight changes are well documented, there was a time from January to roughly July (maybe it still persists) where the API constantly spew 503 errors. This was so bad that I switched to OpenAI API instead. See e.g. https://discuss.ai.google.dev/t/frequent-503-errors-service-...
Cannot say I’ve shared that experience. Requests 504ed 5/6 times during my short trials with curl
Good point. On that note, it may be a bug but I noticed Argon compacting well under the 1M max in Copilot (around 400k) -- when set to 1M context. That may be a Copilot thing but I've never seen that with other models running 1M context. Could be a day-1 bug.
Outside of that, behavior is quite good, a bit slow, and definitely different than both GPTs and Claudes -- which is appreciated.
As a personal anecdote I highly recommend not using the Copilot harness. Even with the same model I found it produced mediocre results against the first class frontier model harnesses.
For my own benchmarks copilot cli uses 1.5x more tokens than opencode.
Consistently buggy you mean. 1 in 10 tasks get stuck in reasoning loops or error out.
>Reliably cheap prices
do you have to pay those cheap prices eventually or can you just accrue an unlimited amount of overdue charges?
I don't think this model supports video yet btw. At least it doesnt work with the API yet.
So cheap that people who pay for pro don't get it as a coding agent in antigravity lol
[flagged]
[flagged]
They also are big contributors to open weight
What's a good LLM model not made by a billionare?