21–30 of 60 posts
Like LLM that only trained on Python 3+, certain frameworks, certain code repos. Then you can use a different model for searching the internet to implement different things to cut down on costs.
Maybe I have no idea what I'm talking about lol
Really looked forward to this release as MiniMax M2.1 is currently my most used model thanks to it being fast, cheap and excellent at tool calling. Whilst I still use Antigravity + Claude for development, I reach for MiniMax first in my AI workflows, GLM for code tasks and Kimi K2.5 when deep English analysis is needed. Not self-hosting yet, but I prefer using Chinese OSS models for AI workflows because of the potent…
> MiniMax first in my AI workflows, GLM for code tasks and Kimi K2.5 Its good to have these models to keep the frontier labs honest! Can i ask if you use the API or a monthly plan? Do the monthly plan throttle/reset ? edit: i agree that MM2.1 most economic, and K2.5 generally the strongest
- $10/mo: 100 prompts / 5 hours
- $20/mo: 300 prompts / 5 hours
- $50/mo: 1000 prompts / 5 hours
[1] https://platform.minimax.io/docs/guides/pricing-coding-plan
Wouldn't it be nice if we have language specific llms that work on average computers. Like LLM that only trained on Python 3+, certain frameworks, certain code repos. Then you can use a different model for searching the internet to implement different things to cut down on costs. Maybe I have no idea what I'm talking about lol
I hope better and cheaper models will be widely available because competition is good for the business. However, I'm more cautious about benchmark claims. MiniMax 2.1 is decent, but one can really not call it smart. The more critical issue is that MiniMax 2 and 2.1 have the strong tendency to reward hacking, often write nonsensical test report while the tests actually failed. And sometimes it changed the existing cod…
That's what I found with some of these LLM models as well. For example I still like to test those models with algorithm problems, and sometimes when they can't actually solve the problem, they will start to hardcode the test cases into the algorithm itself.. Even DeepSeek was doing this at some point, and some of the most recent ones still do this.
It wrote an extensive test suite on just fake data and then said the app is perfectly working as all tests passed.
This is a model that was supposed to match sonnet 4.5 in benchmarks. I don't think sonnet would be that dumb.
I use LLMs a lot to code, but these chinese models don't match anthropic and openai in being able to decide stuff for themselves. They work well if you give them explicit instructions that leaves little for it to mess up, but we are slowly approaching where OpenAI and anthropic models will make the right decisions on their own
We've done some vibe checks on it with OpenHands and it indeed performs roughly as good as Sonnet 4.5.
OSS models are catching up
I hope better and cheaper models will be widely available because competition is good for the business. However, I'm more cautious about benchmark claims. MiniMax 2.1 is decent, but one can really not call it smart. The more critical issue is that MiniMax 2 and 2.1 have the strong tendency to reward hacking, often write nonsensical test report while the tests actually failed. And sometimes it changed the existing cod…
Instead, this one works surprisingly well for the cost: https://openrouter.ai/xiaomi/mimo-v2-flash
Earlier quoted context omitted.
> MiniMax first in my AI workflows, GLM for code tasks and Kimi K2.5 Its good to have these models to keep the frontier labs honest! Can i ask if you use the API or a monthly plan? Do the monthly plan throttle/reset ? edit: i agree that MM2.1 most economic, and K2.5 generally the strongest
Using a coding plan, haven't noticed any throttling and very happy with the performance. They publish the quotas for each of their plans on their website [1]: - $10/mo: 100 prompts / 5 hours - $20/mo: 300 prompts / 5 hours - $50/mo: 1000 prompts / 5 hours [1] https://platform.minimax.io/docs/guides/pricing-coding-plan
Pelican is recognizable but not great, bicycle frame is missing a bar: https://gist.github.com/simonw/61b7953f29a0b7fee1f232f6d9826...
Earlier quoted context omitted.
That's what I found with some of these LLM models as well. For example I still like to test those models with algorithm problems, and sometimes when they can't actually solve the problem, they will start to hardcode the test cases into the algorithm itself.. Even DeepSeek was doing this at some point, and some of the most recent ones still do this.
I have asked GLM4.7 in opencode to make an application to basically filter a couple of spatial datasets hosted at a url I provided it, and instead of trying to download read the dataset, it just read the url, assumed what the datasets were (and got it wrong) is and it's shape (and got it wrong) and the fields (and got it wrong) and just built an application based on vibes that was completely unfixable. It wrote an ex…
Just now it added some stuff to a file starting at L30 and I said "that one line L30 will do remove the rest", it interpreted 'the rest' as the file, and not what it added.