Eni Grand

Enigrand's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 2 days ago

Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Recurrent Half of a Hybrid 27B LLM

Paper • 2609.04098 • Published Sep 3 • 86

upvoted a collection 5 days ago

All QUASAR Models

Collection

All QUASAR checkpoints in one place: 4-bit QAT for Qwen, Gemma and Muse — NVFP4 W4A16/W4A4 for vLLM, Q4_0 GGUF for llama.cpp / Ollama. • 12 items • Updated 22 days ago • 2

upvoted a paper 5 days ago

QUASAR: Lowering the Loss Floor of Quantization-Aware Training with Loss-Aware Reconstruction

Paper • 2608.13966 • Published Aug 14 • 6

upvoted a paper 10 days ago

Disaggregated Quantization: Specializing LLM Prefill and Decode

Paper • 2609.26333 • Published 17 days ago • 93

upvoted a collection 13 days ago

Swift1.5 Flash Next

Collection

Swift Flash Next: the next Swift generation. Weights and quantized builds are added here as they are released. • 6 items • Updated about 14 hours ago • 10

upvoted a collection 14 days ago

Swift 1.5 27B

Collection

Swift 1.5 on Qwen3.8-27B: stronger than Swift 1.0 on agentic and coding tasks, with fewer thinking tokens. BF16 weights and every quant. • 11 items • Updated about 14 hours ago • 23

upvoted a collection 21 days ago

Bonsai-2

Collection

4 items • Updated 21 days ago • 165

upvoted a paper about 1 month ago

Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $5090

Paper • 2608.27370 • Published Aug 27 • 40

upvoted 2 collections about 1 month ago

Muse Glimmer

Collection

Muse Glimmer 30B: multimodal agentic model for local deployment. BF16 weights, GGUF k-quants, ExecuTorch builds, DFlash drafter. • 4 items • Updated Aug 10 • 112

Granite 4.2 Language Models

Collection

Efficient reasoning and thinking language models for multilingual generation, coding, and AI assistant workflows. • 24 items • Updated 24 days ago • 45

upvoted 2 papers about 2 months ago

MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenarios

Paper • 2603.09983 • Published Feb 12 • 4

FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution

Paper • 2608.16157 • Published Aug 17 • 113

upvoted 2 collections about 2 months ago

DFlash 2

Collection

Keep Drafting Parallel • 6 items • Updated Aug 28 • 51

Qwen3.8

Collection

4 items • Updated Aug 13 • 557

upvoted a paper 2 months ago

From RLVR to RLSVR: Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement

Paper • 2607.23802 • Published Jul 26 • 97

upvoted a collection 2 months ago

North Mini Code 1.0

Collection

4 items • Updated Jun 18 • 6

upvoted 3 collections 3 months ago

Gemma 4 QAT Mobile

Collection

4 items • Updated Jul 21 • 52

Agents-A1

Collection

Agents-A1 is a Long-horizon Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. • 12 items • Updated Jul 16 • 45

Bonsai 27B

Collection

10 items • Updated Jul 15 • 248

upvoted a paper 3 months ago

DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation

Paper • 2607.05147 • Published Jul 6 • 51