Qwenputer compiles a RISC-V CPU into Qwen 3.5 4B weights
The experimental release treats Gated DeltaNet state as exact RAM and executes C and MicroPython during forward passes.
Stories per day · 30 days
2 RSS
Relatedagentic-tasks1compiled-weights1inference1linear-attention1long-context1mamba1memory-models1memt1merlin1qwen1qwenputer1risc-v1
8 stories, newest first · one card per story
Stories tagged #gated-deltanet
MERLIN uses key moments to route multi-state linear attention
Reported long-context QA gains are +8.3 and +7.0 points on Gated DeltaNet-1.3B and Mamba-2-780M, with larger LongBench lifts.
MemT-1.0 releases Nano and Flash memory-native models
At 128K history, Nano is about 1.86 times faster and uses 66 times less persistent storage than a full-context KV cache.
HLA-WM: Hybrid Linear Attention for Long-Horizon Video World Models
HLA-WM improves long-horizon video world-model consistency without training while cutting memory use.
HLA: Expressive Hybrid Linear Attention via Chunk-Wise Dynamic Mixing
Hybrid Linear Attention adds query-dependent chunk routing to Gated DeltaNet, lifting long-context scores on Qwen3.5.
LeapQuant: Efficient Linear Attention with Accurate Recurrent State Quantization
LeapQuant nearly matches FP32 quality with training-free 8-bit recurrent-state quantization.
SpectralShift: Effective Context Window Extension of Gated DeltaNet via Spectral Reparameterization
SpectralShift extends Gated DeltaNet context windows via spectral reparameterization of alpha projections, improving long-context capabilities without architectural changes.
LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay
First demonstrated cross-model handoff of persistent recurrent inference state between Qwen3.5 4B and 9B hybrid models without target prefix replay.