# 📆 ThursdAI - Aug 21 - DeepSeek V3.1’s hybrid upset, ByteDance’s 512K Seed-OSS, Nano Banana wizard...

Source: https://www.youtube.com/watch?v=ekjYmxAx_10
Recap page: https://rapidrecap.app/video/ekjYmxAx_10
Generated: 2025-08-28T10:29:20.882+00:00

---
## Quick Overview

The week's open-source AI landscape saw significant releases, with DeepSeek V3.1 emerging as a hybrid reasoning model that matches or surpasses its predecessor R1 in performance while being faster. Other key releases included ByteDance's 36B parameter Seed-OSS with a 512K context window, Coher's Command R reasoning model, and Nvidia's Neotron Nano 9B V2, a mixed Mamba/Transformer model with extended context capabilities. The episode also touched upon the growing trend of agentic models and the challenge of managing diverse agent configuration files, leading to efforts like Open AI's agents.mmd.

**Key Points:**
- DeepSeek V3.1, a hybrid reasoning model, offers comparable or better performance than R1 with reduced thinking tokens, indicating increased efficiency and speed.
- ByteDance launched Seed-OSS, a 36 billion parameter model with a 512K context window, under an Apache 2 license, showing impressive benchmarks against models like GPT-OSS.
- Coher released Command R, a reasoning-focused model with strong performance on agentic benchmarks like BFCL, though it is under a non-commercial research license.
- Nvidia unveiled Neotron Nano 9B V2, a 9 billion parameter model featuring a mixed Mamba/Transformer architecture and compressed to support up to 128K context tokens on a single GPU, with a notable increase in throughput.
- The discussion highlighted a trend towards agentic models across various providers, with an acknowledgment of the challenges in managing multiple agent configuration files, leading to initiatives like Open AI's agents.mmd for unification.
- IBM and NASA collaborated on Surya, an open-source model for solar weather prediction trained on extensive NASA data, and Nvidia released its Neotron Nano 9B V2 with a significant portion of its pre-training data made public.

**Context:** This episode of Thursday Eye (August 21st) covers the latest breaking news in open-source AI, with hosts Alex Walov and Wolf from joined by Non and LJ. The discussion centers on recent model releases that have surprised the community with their capabilities and efficiency. Key players like DeepSeek, ByteDance, Coher, and Nvidia are featured, alongside discussions on emergent trends like hybrid reasoning models and agentic capabilities.

## Detailed Analysis

The week in open-source AI was marked by several high-impact releases. DeepSeek launched V3.1, a hybrid reasoning model that offers performance on par with or exceeding its predecessor, R1, but with significantly fewer thinking tokens, leading to faster responses. ByteDance contributed with Seed-OSS, a 36B parameter model boasting a 512K context window and strong benchmark scores, released under an Apache 2 license. Coher introduced Command R, a reasoning model demonstrating competitive agentic capabilities, though it is restricted to non-commercial use. Nvidia released Neotron Nano 9B V2, notable for its mixed Mamba/Transformer architecture and its ability to handle up to 128K context tokens efficiently. The conversation also touched upon the growing importance of agentic AI, with Open AI's agents.mmd aiming to standardize agent configuration management amidst a proliferation of tools. Other releases mentioned include IBM and NASA's Surya for solar weather prediction and Nvidia's commitment to open-sourcing data for its Neotron models.

### Open Source AI Releases

- DeepSeek V3.1
- ByteDance Seed-OSS
- Coher Command R
- Nvidia Neotron Nano V2
- IBM/NASA Surya

### DeepSeek V3.1 Analysis

- Hybrid reasoning model
- Faster than R1 with fewer tokens
- Improved benchmarks in math, coding, and terminal tasks
- Supports up to 164K context

### ByteDance Seed-OSS Details

- 36B parameters
- 512K context window
- Apache 2 license
- Strong performance on MMLU Pro and GPQA benchmarks

### Coher Command R Overview

- Reasoning model
- Focus on enterprise
- Strong BFCL scores for agentic tasks
- Non-commercial research license

### Nvidia Neotron Nano V2 Features

- 9B parameters
- Mixed Mamba/Transformer architecture
- Up to 128K context on single GPU
- 6.6 trillion tokens of pre-training data released

### Emerging AI Trends

- Rise of agentic models
- Need for unified agent configuration (agents.mmd)
- Hybrid models gaining traction
- Focus on efficiency and speed