natrys

6 ms·

1.
by natrys 6d ago
According to an old interview, apparently they were always interested in AI. But finance is just where they had their first success. > Many of High-Flyer's original team members worked on AI. Back then, we tried a lot of fields befo
2.
by natrys 19d ago
I think we don't have a good draft model for better speculative decoding yet (e.g. DFlash 2). Once we do, it will be faster.
3.
by natrys 22d ago
Why not? It's not really competing in the same size class. Besides, as they explicitly wrote here, the main goal for this release is not performance, rather to serve as a reference for inference runtimes about what to implement. So tha
4.
by natrys 1mo ago
For me, flash 0731 was much better in omp/opencode than in Pi. Anyway, it might be so that they are rolling out deployment. There haven't been an official announcement post yet (this submission is a link to openrouter). Some peo
5.
by natrys 1mo ago
There is an old rule of thumb that says the quality of an MoE is equivalent to a dense model with the geometric mean of its total and active parameters. So, for example, the Qwen3.6 would be equivalent to a dense model with approximately sq
6.
by natrys 1mo ago
I think this one requires a bit of strong prompting. I am also normally a Pi user, but my experience in OpenCode with this model has been drastically better than in Pi, where it overthinks a lot and gets distracted by random things. It migh
7.
by natrys 1mo ago
So your cost would be $40.68 with another provider that has one less zero in the cache hit price.
8.
by natrys 2mo ago
At least the SSM layers should have relative PE built in via recurrence, in a hybrid attention model like this.
9.
by natrys 2mo ago
They can and almost certainly are doing similarly impressive engineering works internally. They just aren't in any hurry to forward those cost savings to you.
10.
by natrys 2mo ago
If anecdote is data, then here's another point: https://nitter.net/synthwavedd/status/2077537805715005724#m (As an aside, I don't know how it was professional of Arena to unmask an unreleased cloaked mod
11.
by natrys 2mo ago
Some official benchmark numbers posted in Chinese social media (I am sure they will publish an English blogpost later too): https://mp.weixin.qq.com/s/V4xhEIy8xDXSMDPrPkmUAQ Generally looks like a Sol/Fable tier m
12.
by natrys 3mo ago
These things enormously benefit from economies of scale. I am fairly certain their margins might be low but they don't actually sell API at loss, however that doesn't mean your cost footprint would be anywhere as low.
13.
by natrys 3mo ago
No I think uv is to python what opam is to ocaml, it's mostly a package/dependency manager. Superficially, both uv and dune are also project runners. But dune is mainly a build tool, most important things dune does such as pre-pro
14.
by natrys 3mo ago
It seems frontier, on the balance, would rather lose that segment of he market than lower the API price. They are getting the bag in the enterprise segment, those clients aren't ditching them for DeepSeek. As for other segments, high A
15.
by natrys 5mo ago
Yes it was good for its time, but 10 months old now which is a long time ago in this space. It was also a fine-tune (albeit a good one) of Qwen-2.5 72B. I wish they did more smaller models. Kimi Linear doesn't really count, it was more
16.
by natrys 7mo ago
I think tree-sitter's relationship with JavaScript is entirely syntactic. You don't need any JS runtime installed to write grammars, because technically tree-sitter CLI already has a JS runtime included and using that it convert
17.
DeepSeek Engram: Conditional Memory via Scalable Lookup [pdf](github.com)
6 points by natrys 8mo ago|2 comments
18.
by natrys 9mo ago
Very impressive demo. From VM curation to vibe coding something running on port 8000 in Shelley just worked in minutes. I imagine quite a few technically impressive things happening under the hood, would be interested in reading more about
19.
by natrys 9mo ago
That's the Kimi K2 Thinking, this post seems to be talking about original Kimi K2 Instruct though, I don't think INT4 QAT (quantization aware training) version was released for this.
20.
by natrys 10mo ago
I am going to try and stick with Prolog as much as I can this year. Plenty of problems involve a lot of parsing and searching, both could be expressed declaratively in Prolog and it just works (though you do have to keep the execution model
21.
by natrys 10mo ago
Well they do that too: https://huggingface.co/deepseek-ai/DeepSeek-Prover-V2-671B But I suppose the bigger goal remains improving their language model, and this was an experimentation born from that. These works are
22.
by natrys 10mo ago
It can hardly be called resistance to improvement, when everyone do improve it - just in their own ways. The default isn't some fashion statement, some aesthete that's objectively good (though I am sure some people do subjectively
23.
by natrys 11mo ago
Wasn't aware of user-var-changed, cool write-up! I had used urxvt forever before and the simple solution that works (even for ssh e.g.) is to ring the terminal bell, and urxvt just sets the window urgency hint upon that. I just do that
24.
by natrys 11mo ago
Yes it seems the binaries are here: https://ferron.sh/download I will say that though, it's probably not rational to be okay with blindly running some opaque binary from a website, but then flip out when it comes to ru
25.
by natrys 11mo ago
Qwen's max series had always been closed weight, it's not a policy change like you are alluding. What exactly is Huawei's flagship series anyway? Because their PanGu line is open-weight, but Huawei is as of yet not in the
26.
Qwen3-VL(qwen.ai)
434 points by natrys 1y ago|160 comments
27.
by natrys 1y ago
Models: - https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Thinking - https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Instruct
28.
by natrys 1y ago
If you mean the bit about refusal from other models, then sure here is another run with same result: https://i.postimg.cc/6tT3m5mL/screen.png Note I am using direct API to avoid triggering separate guardrail models typ
29.
by natrys 1y ago
You are obviously factually correct, I reproduced the same refusal - so consider this not as an attack on your claim. But a quick google search reveals that Falun Gong is an outlawed organization/movement in China. I did a "s&#x
30.
by natrys 1y ago
I agree with you, therefore I am pretty sure you meant to reply to the parent I was also replying to.

More ›