Mamba: Linear-Time Sequence Modeling with Selective State Spaces
by Carnegie Mellon University / Together AI · free · Last verified 2026-03-17
Mamba is a novel sequence modeling architecture based on structured state space models (SSMs). It introduces a selection mechanism that allows the model to selectively propagate or forget information based on the input, overcoming a key limitation of previous SSMs. This enables Mamba to achieve Transformer-level performance with linear time complexity and significantly faster inference.
https://arxiv.org/abs/2312.00752 ↗B
B—Above Average
Adoption: B+Quality: A+Freshness: ACitations: BEngagement: F
Specifications
- License
- Apache 2.0
- Pricing
- free
- Capabilities
- sequence-modeling, linear-time-complexity, fast-autoregressive-inference, selective-state-spaces, long-context-handling, causal-language-modeling, efficient-hardware-aware-training, attention-free-architecture, recurrent-and-parallel-computation-modes
- Integrations
- Use Cases
- [object Object], [object Object], [object Object], [object Object]
- API Available
- No
- Tags
- mamba, state-space-model, ssm, linear-time, selective-state-space, recurrence, transformer-alternative, long-context, sequence-model, efficient-inference, ai-architecture
- Added
- 2026-03-17
- Completeness
- 0.95%
Index Score
63.8Adoption
72
Quality
90
Freshness
80
Citations
68
Engagement
0