Hacker News

points by A_D_E_P_T 3 days ago | 0 comments

This is awesome. ERNIE seems more powerful than Qwen3.8-27B and Deepseek-v4-Flash, which AFAIK are currently the best base models... it'll be very interesting to experiment with when it hits OpenCode. Base models are also excellent for businesses who run models locally, as they can be trained in various house rules and house styles.

[–] input_sh 3 days ago

They're comparing it to Deepseek-v4-Flash, not 4.1 and not the 4.1 1017 update, so essentially two versions behind the latest version of Deepseek Flash.

[–] Tepix 3 days ago

DeepSeek 4.1 Flash is twice the size of this and difficult to run locally. Makes more sense to compare it to DS4F.

[–] segmondy 2 days ago

Do you understand that this is a base model? It means you are supposed to take it and post train it. It's much smaller than DeepSeek-v4-Flash, that's already a plus, you compare base to base. 4.1 and 4.1-1017 are trained from the base model.

[–] input_sh 2 days ago

I understand why you might think so, as calling it v4.1 does imply that, but no, 4.1 was built from scratch.

Different architecture, v4 is text-only while v4.1 is multimodal and so on.

[–] lcampbell 2 days ago

> v4 is text-only

To be a pedant, DeepSeek-V4-Flash-Vision-Exp was a further checkpoint of the V4.0 lineage that supported image input. It lost a few benchmark points compared to the previous 0731 checkpoint so a lot of people skipped it. Completely unrelated to V4.1 though.

I wish labs used more reasonable naming schemes.

[–] irthomasthomas 3 days ago

Qwen3.8-27B has no base model version in the wild that I can see?

[–] speedgoose 3 days ago

Isn’t the latest Qwen 3 public base model… Qwen 3?

[–] stymaar 2 days ago

Despite the naming Qwen3 and Qwen3.5+ have very little in common in terms of architecture. (And neither does 3.8-Flash-next …)

[–] halJordan 2 days ago

Baidu, Tencent, all the second rate Chinese labs' models are all benchmaxxed to hell. The main goal of these releases is to show benchmarks to their govt minders

[–] A_D_E_P_T 2 days ago

Have you tried Xiaomi's MiMo v2.6? It's remarkably good -- and very weird/idiosyncratic. (For instance, it can compose MIDI-style music directly via the chat app, and it follows instructions much more closely than any other music model. AFAIK nothing else does this.)

I haven't had much experience with ERNIE, but I've been using various different (Chinese) base models of late, and I'm excited to try it.