superoo7

@not_qz @Alibaba_Qwen gif or jif?

@superoo7 @Alibaba_Qwen Is it qwen or Qwen

Excited to share I've joined the @Alibaba_Qwen Ambassador program 🎉 Looking forward to running events and helping grow the Qwen ecosystem Always happy to support great open-source projects 🙌

@0xDamien Come join me brother

Huge mispricing $BABA Fuck yo China risk

@Zalam195 @Alibaba_Qwen yeap doing my part

@superoo7 @Alibaba_Qwen $eval is undervalued, need visibility.

Big thanks to @HaaaaaaydenH and the @Alibaba_Qwen team for the warm welcome 🙏

@dphnAI @AskVenice @0xSUPERGEMMA @jangqai @bankrbot @Eval_Engine @Chromia @virtuals_io @base cc @AIonBase_

The rest of the wave worth watching: $POD @dphnAI (launched on @AskVenice) $SUPERGEMMA @0xSUPERGEMMA from @jangqai (launched on @bankrbot) $EVAL from @Eval_Engine & @Chromia (launched on @virtuals_io) All open-weight, all on @Base 👀

This is happening faster than I expected: @dealignai $DEAL just launched, newest of the "uncensored-model-token wave" Open weights → real HF downloads → token on @base $POD $DEAL $SUPERGEMMA $EVAL

@PengsBingBing @Eval_Engine @dphnAI @AskVenice @0xSUPERGEMMA @jangqai @bankrbot @Chromia @virtuals_io @base 👀

Thanks @XiaomiMiMo for supporting @Eval_Engine, we got 200M tokens to spend for improving it 🔥

96hrs since Unbound went live on @Eval_Engine Total of 6k downloads across various Unbound models running fully offline No servers, no cloud subscription Chat with Unbound - unbound.evalengine.ai chromaway:native $EVAL

@RRekt46321 @Eval_Engine @Chromia we trained it with some chromia dataset

@Eval_Engine Are you guys using @chromia with it?

@huggingface @base @dphnAI @0xSUPERGEMMA @Eval_Engine @jun_song Unbound from Eval Engine huggingface.co/collections/ev…

@huggingface @base @dphnAI @0xSUPERGEMMA @Eval_Engine Supergemma from @jun_song

More and more open-weight uncensored models on @Huggingface are launching tokens on @Base 👀 @dphnAI $POD — Dolphin series @0xSUPERGEMMA $SUPERGEMMA — SuperGemma (Gemma fine-tune) @eval_engine $EVAL — Unbound model

I've been following Qwopus's model for a while, great work! If you're into fine tuning your own model, can also take a look at their publication, truly a rare 💎 github.com/R6410418/Jackr…

@KyleHessling1 aite will take a look into this as well 🙏 thanks for the great work!

@superoo7 I believe if you use the Qwen3.6-27B-DFlash drafter it should work! I will have to take it for a spin this weekend to verify, but it should just work as if you were using the base! If the acceptance rate is bad, we could train a drafter for it!

BREAKING! Qwopus 3.6 27B is LIVE! Thank you for your patience on this one, but I believe you'll find the wait was worth it! We've benchmarked this thing up and down, verified that it holds at least a 75.25% (152/202) in the initial 202 SWE bench solves. Not a full run of 500, but it shows the agentic coding quality from the original 27B is retained while adding all of the additional Qwopus benefits across many domains. As always, Jackrong is absolutely cooking here! COT quality has improved significantly through the inversion techniques from our Negentropy proof of concept. It also went through thorough curriculum training. You can check out the MMLU pro benchmarks on the model card, but it improved a whopping 10 points over the base model in physics, as well as meaningful jumps in Chemistry, business, and computer science. However, the best part is that I was able to build an entire survival shooter game using this local model entirely. I genuinely was blown away by the results, which you can play right now on my HF space (link in comments below). "Qwopus Commander" was completed in 9 turns of Qwopus 3.6! To test the new long context training, I made it re-output the entire 3000+ line program each turn, and it would make fixes and add features that I requested in large prompts, while perfectly replicating the entire rest of the game from context. What's more is that I did it all at Q8 KV cache quantization, and never had an issue over the entire 303k token run! IMPORTANT: Run it at --temp 0.75 to 1. Mess with it in that range for your use case. Higher temp actually lets the fine-tune shine and be exploratory and is also more stable. Swe Bench was run at temp 1, the game was built mostly at 0.8! We're so blessed to have all of you here and using the models! The support means so much! Please let me know what you build with it in the comments! Or if you have any issues getting it up and running, I will try my best to get back to you! Looking forward to seeing what you legends produce with it this weekend! huggingface.co/Jackrong/Qwopu…

wllama -> llama.cpp’s WebGPU expecting some performance boost on our Unbound model demo soon 👀