Timothy. Z

AI @Wordware. Building proactive AI agents for teamwork.
Also, independent researches, post training experiments with local models

[Timothy. Z @TimothyZ77]

@yosihideganbare

[Timothy. Z @TimothyZ77]

If GPU price goes down, no one will remember what happened on Nov 8th

[NVIDIA Newsroom @nvidianewsroom]

We just issued a press release from our board of directors. Read the release: https://t.co/W9yXWJzbkk

[Timothy. Z @TimothyZ77]

@ykni0245

[Timothy. Z @TimothyZ77]

LMAO a system prompt

[169Pi @169Pi_ai]

This Diwali, a little more light from India's AI research community onto the global AI landscape.

Introducing the latest benchmark results from Alpie Seed, our 9B multimodal reasoning model developed at @169Pi_ai Research.

We believe the next generation of AI will be shaped not only by models that scale bigger, but by the research and engineering that make smaller models increasingly capable.

We put Alpie Seed through a range of demanding evaluations across reasoning, scientific knowledge, mathematics, coding and software engineering.

Here are some of the results:

→ 82.54% on MMLU-Pro
→ 84.32% on GPQA Diamond
→ 72.99% on LiveCodeBench v6
→ 61.20% on SWE-Bench Verified

The comparison highlights some interesting results.

Alpie Seed records the highest GPQA Diamond score among the models shown. On MMLU-Pro, it performs competitively with larger models, while its LiveCodeBench result is within striking distance of substantially larger systems.

Its SWE-Bench Verified result of 61.20% is another important milestone, reflecting its ability to work through real-world software engineering tasks. This evaluation used mini-SWE-agent without retries.

Beyond benchmark accuracy, we also examined inference efficiency.

In a separate profiling evaluation, Alpie Seed recorded:

→ 36.66 ms time to first token
→ 41.37 tokens/s generation speed
→ 9,796.22 tokens/s prompt processing speed

These measurements come from a separate profiling setup and should not be confused with the comparative benchmark results.

What makes this work meaningful to us is the process behind it: developing a compact model through supervised fine-tuning and reinforcement learning, evaluating it across diverse tasks, and making the resulting weights accessible to the wider community.

The comparison includes published results from other model developers. These evaluations were not rerun under a common harness, so differences in methodology and configuration should be considered when interpreting the numbers.

Built in India. Developed for a global research and developer community.

This is Alpie Seed. And this is one step in a much larger journey for 169Pi Research.

Explore the model:

Hugging Face: https://t.co/Jpujzrnnoy
GGUF: https://t.co/tD332KqcCo
Ollama: https://t.co/7GqHxQQsvF
Kaggle: https://t.co/9ZHdYgSjBB

Happy Diwali from everyone at 169Pi.

[Timothy. Z @TimothyZ77]

？

[Congresswoman Yassamin Ansari @RepYassAnsari]

Top of the Democratic agenda on day 1 needs to be a pause on advanced AI development, regulation of this out of control industry, and a ban on artificial superintelligence.

[Timothy. Z @TimothyZ77]

@Nightmux ?

[Timothy. Z @TimothyZ77]

🤨

[Remarks @remarks]

JUST IN: 🇺🇸 Missing election machine memory card discovered in Georgia flips state legislative race from Democrat to Republican.

[Timothy. Z @TimothyZ77]

@Teknium And this question intensifies after reading the dataset, thanks for the transparency though

[Timothy. Z @TimothyZ77]

Is western values no propaganda or western propaganda?

[Nous Research @NousResearch]

Introducing Arete-27B: Qwen3.8-27B, post-trained to reinforce Western values.

Chinese open models are currently the strongest available, but they ship with state-aligned behavior in the weights. We built an 803-prompt eval across seven value categories, including freedom of expression, democracy, and privacy, and measured how often a model adopts China’s party line.

After fine-tuning on roughly 53 million tokens, Arete-27B's propaganda rate fell from 22% to 1%, in line with the Western models we tested. Its Hermes Bench score shows no significant performance tradeoffs.

The values in an open model are one fine-tune away.

Our eval, weights, and training data are available now along with the full blog post: https://t.co/iAOANQTVZ6

[Timothy. Z @TimothyZ77]

@zainhas Open or not is likely inevitable on student models that went through too many rounds of post training

[Timothy. Z @TimothyZ77]

@serious_mehta Rather than benchmaxxed, it’s more so the coding tasks look saturated

[0xSero @0xSero]

Caste system is looking to be true

[Justin Lo @loresjpg]

Nice DeepSWE score! Does it survive a lowercase filename?

Introducing MaxxBench: 89 DeepSWE tasks with small changes to the instructions or starting code. Same engineering problem, slightly different requirement.

Models clustered around 74% on DeepSWE end up 23 pp apart.

Good at coding, or good at DeepSWE? 🧵

[Timothy. Z @TimothyZ77]

@follistatindev Coding tasks look saturated at frontier? But interesting

[Timothy. Z @TimothyZ77]

@ThePhysicsMemes What are you scheming on earth

[Jameson Hodge @jamesonhodge]

I only make serious software

[Timothy. Z @TimothyZ77]

@TeamVitalityVAL

[Timothy. Z @TimothyZ77]

@yuvals41 Hey

[Timothy. Z @TimothyZ77]

@iam_sid26 Hi

[Timothy. Z @TimothyZ77]

@BekaluAdane1997 Hello