Pith. sign in

REVIEW 7 cited by

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.14458 v2 pith:CKWL7ZDS submitted 2025-02-20 cs.LG cs.AI

Llamba: Scaling Distilled Recurrent Models for Efficient Language Processing

classification cs.LG cs.AI
keywords llambamodelslanguagedistilledefficiencyefficientperformancerecurrent
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We introduce Llamba, a family of efficient recurrent language models distilled from Llama-3.x into the Mamba architecture. The series includes Llamba-1B, Llamba-3B, and Llamba-8B, which achieve higher inference throughput and handle significantly larger batch sizes than Transformer-based models while maintaining comparable benchmark performance. Furthermore, Llamba demonstrates the effectiveness of cross-architecture distillation using MOHAWK (Bick et al., 2024), achieving these results with less than 0.1% of the training data typically used for models of similar size. To take full advantage of their efficiency, we provide an optimized implementation of Llamba for resource-constrained devices such as smartphones and edge platforms, offering a practical and memory-efficient alternative to Transformers. Overall, Llamba improves the tradeoff between speed, memory efficiency, and performance, making high-quality language models more accessible.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Morphing into Hybrid Attention Models

    cs.CL 2026-06 unverdicted novelty 7.0

    FlashMorph formulates hybrid layer selection as budget-constrained optimization, trains per-layer gates on synthetic retrieval data with linearization regularization, then discretizes and distills to produce efficient...

  2. The Key to Going Linear: Analysis-Driven Transformer Linearization

    cs.LG 2026-07 conditional novelty 6.0

    Delta-rule linear attention faithfully approximates softmax attention through key-dependent rank-1 projections, enabling efficient post-hoc linearization of LLMs up to 32B parameters.

  3. Contribution Weights: A Geometrical Analysis of Self-Attention Transformers

    cs.LG 2026-05 unverdicted novelty 6.0

    Contribution Weights combine attention, value magnitude, and directional alignment to measure token influence more faithfully than attention alone, and show attention sinks actively suppress information via a convex s...

  4. Long-Context Aware Upcycling: A New Frontier for Hybrid LLM Scaling

    cs.CL 2026-04 unverdicted novelty 6.0

    HyLo upcycles Transformer LLMs into hybrids with MLA and Mamba2/Gated DeltaNet blocks via staged training and distillation, extending context to 2M tokens and outperforming prior upcycled hybrids on long-context benchmarks.

  5. Attention to Mamba: A Recipe for Cross-Architecture Distillation

    cs.CL 2026-04 unverdicted novelty 6.0

    A two-stage distillation recipe converts a Pythia-1B Transformer into a Mamba model that preserves performance with perplexity 14.11 versus the teacher's 13.86.

  6. MAR: Efficient Large Language Models via Module-aware Architecture Refinement

    cs.AI 2026-01 unverdicted novelty 5.0

    MAR integrates SSMs and sparsification with new ATMN neurons and SBDS distillation to produce efficient LLMs that match dense-model performance at substantially lower inference energy.

  7. SpikingMamba: Towards Energy-Efficient Large Language Models via Knowledge Distillation from Mamba

    cs.NE 2025-10 unverdicted novelty 5.0

    SpikingMamba distills Mamba into an SNN LLM achieving 4.76x energy savings with a 4.78% zero-shot accuracy gap that narrows to 2.23% after RL.