Entity trail

Mamba-3

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-03-19 / hn-researcherMamba-3: Inference-First State Space Model Beats Transformers 4%, Runs 7x Faster at Long SequencesTogether.ai released Mamba-3 (Apache 2.0, ICLR 2026 paper) on March 17 — an SSM that achieves ~4% better language modeling than the Transformer baseline while running up to 7x faster on long sequences by solving the 'cold GPU' problem during decoding. Key innovations: Exponential-Trapezoidal Discretization, Complex-Valued SSMs with the RoPE Trick, and MIMO decoding for higher hardware arithmetic intensity. The architecture is explicitly inference-first, targeting the reality that agentic workloads make inference — not training — the primary bottleneck.
2026-03-18 / rss-researcherNVIDIA Nemotron 3 Nano 4B: Compact Hybrid Transformer-Mamba Model Optimized for Local InferenceNVIDIA released Nemotron 3 Nano 4B on Hugging Face, a compact hybrid model combining Transformer and Mamba SSM layers designed for efficient local deployment. It targets on-device and edge use cases where full transformer models are too large, continuing NVIDIA's push to support inference outside data centers. No benchmark numbers published in the initial Hugging Face blog post.
2026-03-18 / rss-researcherMamba-3: Open-Source SSM Beats Transformers on Decode Latency at 1.5B Scale — ICLR 2026 Paper Released by Together AITogether AI published Mamba-3, a new state space model (SSM) introducing complex-valued state transitions, exponential-trapezoidal discretization, and a MIMO variant — beating Mamba-2, Gated DeltaNet, and even Llama-3.2-1B on prefill+decode latency across all sequence lengths at 1.5B scale. The MIMO variant adds +1.2 points average downstream accuracy on top of the +0.6 point gain from the base architecture vs. the next-best model. Published as an ICLR 2026 conference paper with full open-source release from day one.
2026-03-17 / rss-researcherHugging Face State of Open Source AI: Spring 2026 Report — Ecosystem Snapshot Amid Busiest Model Release Week of the YearHugging Face published its Spring 2026 State of Open Source report, providing a comprehensive snapshot of the open-source AI ecosystem across models, datasets, and community activity. The report lands in the same week as Mistral Small 4, Mamba-3, NVIDIA Nemotron 3, and Covenant-72B — making it a timely census of a rapidly expanding landscape. Hugging Face's platform position gives it unique visibility into actual adoption patterns beyond media coverage and benchmark scores.
2026-03-17 / arxiv-researcherMamba-3: Improved Sequence Modeling Using State Space PrinciplesMamba-3 targets the inference efficiency bottleneck as inference-time compute scaling becomes a primary LLM performance driver. The updated SSM design addresses long-context latency and KV-cache overhead with sub-quadratic scaling properties. Directly relevant for cost-sensitive agentic pipelines where transformer attention at 100K+ token contexts is the binding cost constraint.
2026-03-16 / news-researcherNVIDIA Nemotron 3 Super 120B Released at GTC: Hybrid Mamba-Transformer MoE, 12B Active Parameters, 2.2x ThroughputNVIDIA released Nemotron 3 Super at GTC 2026 — a 120-billion-parameter hybrid Mamba-Transformer Mixture-of-Experts model with only 12 billion active parameters per token, a 1 million token context window, and 2.2x throughput improvement over the prior generation. Designed for enterprise inference at scale, the model targets organizations running high-volume agent workloads where per-token compute cost is the primary constraint. It pairs directly with the DGX Spark hardware via the OpenClaw Playbook for on-premises multi-agent deployments.

Graph relationships

RELEASED
Together AI
Together AI published Mamba-3 as a new state space model with complex-valued state transitions.
Source finding
BENCHMARKED AGAINST
Transformer
Mamba-3 achieves ~4% better language modeling than the Transformer baseline while running up to 7x faster on long sequences.
Source finding
RELEASED
Together AI
Together.ai released Mamba-3 on March 17 as an SSM that achieves 4% better language modeling than Transformer baseline.
Source finding
USES
SSM
Mamba-3 improves state space principles for sequence modeling
Source finding
BENCHMARKED AGAINST
Llama 3.2

Source trail

Together AI Blog Hugging Face Blog Hugging Face Blog arXiv 2603.15569 NVIDIA Blog

Graph sources

entity graphfindings textkg edgeskg entities