Hackernews posts about Mamba/Transformer

1. 01
   

   new'Western Qwen': IBM Wows with Granite 4 LLM Launch and Hybrid Mamba/Transformerventurebeat.com

   ▲ 832bluesc400d💬 25savesaved
2. 02
   

   newNemotron 3 Ultra: Open Moe Hybrid Mamba-Transformer for Agentic Reasoning [pdf]research.nvidia.com

   ▲ 23victormustar155d💬 2savesaved
3. 03
   

   newNemotron-H: A Family of Accurate, Efficient Hybrid Mamba-Transformer Modelsresearch.nvidia.com

   ▲ 3T-A590ddiscusssavesaved
4. 04
   

   newNemotron 3 Super: An Open Hybrid Mamba-Transformer Moe for Agentic Reasoningdeveloper.nvidia.com

   ▲ 2pr337h4m240ddiscusssavesaved
5. 05
   

   newMamba (Transformer Alternative): The Future of LLMs?lazyprogrammer.me

   ▲ 2JPLeRouzic689ddiscusssavesaved
6. 06
   

   newNemotron-H: A Family of Accurate, Efficient Hybrid Mamba-Transformer Modelsresearch.nvidia.com

   ▲ 1Bluestein45ddiscusssavesaved
7. 07
   

   newNvidia debuts Nemotron 3 with hybrid MoE and Mamba-Transformerventurebeat.com

   ▲ 1_____k326ddiscusssavesaved
8. 08
   

   newHunyuan-TurboS: the first ultra-large Hybrid-Transformer-Mamba MoE modeltwitter.com

   ▲ 1tosh543ddiscusssavesaved
9. 09
   

   newWhat would you do if you have AI software that may be transformers alternative?

   ▲ 3adinhitlore229d💬 5savesaved
10. 10
    

    newNvidia: Nemotron 3 Superdeveloper.nvidia.com

    ▲ 2tosh156ddiscusssavesaved
11. 11
    

    newGenesis Open Source Embodied AGI Simulation, Rust (Mamba-3, Not Transformers)

    ▲ 2RGBra319ddiscusssavesaved
12. 12
    

    newSessa: An alternative to Transformers and Mamba for long-context LLMsgithub.com

    ▲ 1guyfloki197ddiscusssavesaved
13. 13
    

    newMamba 3 matches Transformer performance at reduced latencyventurebeat.com

    ▲ 1ryan_j_naughton234ddiscusssavesaved
14. 14
    

    newMamba-Shedder: Post-Transformer Compression for Efficient SSMsarxiv.org

    ▲ 1fovc626ddiscusssavesaved
15. 15
    

    newShow HN: Linear RNN/Reservoir hybrid generative model, one C file (no deps.)raw.githubusercontent.com

    ▲ 7adinhitlore211d💬 2savesaved
16. 16
    

    newAsk HN: Release Path for 'Transformers Alternatives'?

    ▲ 4adinhitlore222d💬 1savesaved
17. 17
    

    newTCNs as Alternative to Transformers?

    ▲ 6adinhitlore159d💬 3savesaved
18. 18
    

    newShow HN: New Cartesia Text-to-Speech Modelwww.cartesia.ai

    ▲ 10cartesia694ddiscusssavesaved
19. 19
    

    newAura – Detecting Fake Cell Towers with RF Fingerprinting AI

    ▲ 8sadpig70417ddiscusssavesaved
20. 20
    

    newComputer use agent with RL training for day trading?

    ▲ 1iiTsEddy547d💬 1savesaved
21. 21
    

    newDeltaProduct: Improving State-Tracking in Linear RNNs via Householder Products

    ▲ 1jul8234588ddiscusssavesaved
22. 22
    

    newShow HN: Mamba2-Jax; Mamba2 implemented in pure Jax/Flaxgithub.com

    ▲ 2cosmonaught1337349ddiscusssavesaved
23. 23
    

    newBamba: An open-source LLM that crosses a transformer with an SSMresearch.ibm.com

    ▲ 207shallow-mind556d💬 69savesaved

· edit