--- title: README emoji: 📐 colorFrom: indigo colorTo: blue sdk: static pinned: false --- # SketchSSM calibrations [SketchSSM](https://github.com/SNU-ARC/SketchSSM) speeds up the decode of hybrid linear-attention models: the recurrent state is updated exactly once per window, and the other decode steps read a compact per-request sketch of it. Each repository here holds the calibration file (`calibration.pt`) for one model checkpoint; it contains no model weights. ```bash vllm serve --sketchssm SketchSSM/ \ --sketchssm-mean-rank 8 --mamba-ssm-cache-dtype float32 ``` Use the calibration of the exact checkpoint (precision included) you serve. To calibrate another model, see the [calibration guide](https://github.com/SNU-ARC/SketchSSM/tree/main/sketchssm/calibration). ## Mamba-2 (NemotronH) [Collection](https://huggingface.co/collections/SketchSSM/sketchssm-mamba-2-nemotronh-6acc6b2da0d63e6bb92377d3) | Calibration | Base checkpoint | Weights | | --- | --- | --- | | [`SketchSSM/Nemotron-3-Nano-30B-A3B-Base-BF16`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-30B-A3B-Base-BF16) | [`nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-Base-BF16) | BF16 | | [`SketchSSM/Nemotron-3-Nano-30B-A3B-BF16`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-30B-A3B-BF16) | [`nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16) | BF16 | | [`SketchSSM/Nemotron-3-Nano-30B-A3B-FP8`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-30B-A3B-FP8) | [`nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8) | FP8 | | [`SketchSSM/Nemotron-3-Nano-30B-A3B-NVFP4`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-30B-A3B-NVFP4) | [`nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4) | NVFP4 | | [`SketchSSM/Nemotron-3-Nano-4B-BF16`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-4B-BF16) | [`nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16) | BF16 | | [`SketchSSM/Nemotron-3-Nano-4B-FP8`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-4B-FP8) | [`nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-4B-FP8) | FP8 | | [`SketchSSM/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8`](https://huggingface.co/SketchSSM/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8) | [`nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8`](https://huggingface.co/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8) | FP8 | | [`SketchSSM/Nemotron-3-Super-NVFP4`](https://huggingface.co/SketchSSM/Nemotron-3-Super-NVFP4) | [`nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4) | NVFP4 | | [`SketchSSM/Nemotron-3.5-Lightning-30B-A3B-Base-BF16`](https://huggingface.co/SketchSSM/Nemotron-3.5-Lightning-30B-A3B-Base-BF16) | [`nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Base-BF16) | BF16 | | [`SketchSSM/Nemotron-3.5-Lightning-30B-A3B-BF16`](https://huggingface.co/SketchSSM/Nemotron-3.5-Lightning-30B-A3B-BF16) | [`nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16) | BF16 | | [`SketchSSM/Nemotron-Cascade-2-30B-A3B-BF16`](https://huggingface.co/SketchSSM/Nemotron-Cascade-2-30B-A3B-BF16) | [`nvidia/Nemotron-Cascade-2-30B-A3B`](https://huggingface.co/nvidia/Nemotron-Cascade-2-30B-A3B) | BF16 | | [`SketchSSM/Nemotron-H-4B-Base-8K-BF16`](https://huggingface.co/SketchSSM/Nemotron-H-4B-Base-8K-BF16) | [`nvidia/Nemotron-H-4B-Base-8K`](https://huggingface.co/nvidia/Nemotron-H-4B-Base-8K) | BF16 | | [`SketchSSM/Nemotron-Labs-3-Elastic-30B-A3B-BF16`](https://huggingface.co/SketchSSM/Nemotron-Labs-3-Elastic-30B-A3B-BF16) | [`nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-BF16) | BF16 | | [`SketchSSM/Nemotron-Labs-3-Elastic-30B-A3B-FP8`](https://huggingface.co/SketchSSM/Nemotron-Labs-3-Elastic-30B-A3B-FP8) | [`nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-FP8`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Elastic-30B-A3B-FP8) | FP8 | | [`SketchSSM/Nemotron-Nano-12B-v2-Base-BF16`](https://huggingface.co/SketchSSM/Nemotron-Nano-12B-v2-Base-BF16) | [`nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2-Base) | BF16 | | [`SketchSSM/Nemotron-Nano-12B-v2-BF16`](https://huggingface.co/SketchSSM/Nemotron-Nano-12B-v2-BF16) | [`nvidia/NVIDIA-Nemotron-Nano-12B-v2`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2) | BF16 | | [`SketchSSM/Nemotron-Nano-12B-v2-VL-FP8`](https://huggingface.co/SketchSSM/Nemotron-Nano-12B-v2-VL-FP8) | [`nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-FP8`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-FP8) | FP8 | | [`SketchSSM/Nemotron-Nano-12B-v2-VL-NVFP4-QAD`](https://huggingface.co/SketchSSM/Nemotron-Nano-12B-v2-VL-NVFP4-QAD) | [`nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-NVFP4-QAD`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-NVFP4-QAD) | NVFP4 | | [`SketchSSM/Nemotron-Nano-9B-v2-Base-BF16`](https://huggingface.co/SketchSSM/Nemotron-Nano-9B-v2-Base-BF16) | [`nvidia/NVIDIA-Nemotron-Nano-9B-v2-Base`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2-Base) | BF16 | | [`SketchSSM/Nemotron-Nano-9B-v2-BF16`](https://huggingface.co/SketchSSM/Nemotron-Nano-9B-v2-BF16) | [`nvidia/NVIDIA-Nemotron-Nano-9B-v2`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2) | BF16 | | [`SketchSSM/Nemotron-Nano-9B-v2-FP8`](https://huggingface.co/SketchSSM/Nemotron-Nano-9B-v2-FP8) | [`nvidia/NVIDIA-Nemotron-Nano-9B-v2-FP8`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2-FP8) | FP8 | | [`SketchSSM/Nemotron-Nano-9B-v2-Japanese-BF16`](https://huggingface.co/SketchSSM/Nemotron-Nano-9B-v2-Japanese-BF16) | [`nvidia/NVIDIA-Nemotron-Nano-9B-v2-Japanese`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2-Japanese) | BF16 | | [`SketchSSM/Nemotron-Nano-9B-v2-NVFP4`](https://huggingface.co/SketchSSM/Nemotron-Nano-9B-v2-NVFP4) | [`nvidia/NVIDIA-Nemotron-Nano-9B-v2-NVFP4`](https://huggingface.co/nvidia/NVIDIA-Nemotron-Nano-9B-v2-NVFP4) | NVFP4 | ## Gated DeltaNet [Collection](https://huggingface.co/collections/SketchSSM/sketchssm-gated-deltanet-6acc6b2d6fa844eabebbf61a) | Calibration | Base checkpoint | Weights | | --- | --- | --- | | [`SketchSSM/Qwen3.5-0.8B-Base-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-0.8B-Base-BF16) | [`Qwen/Qwen3.5-0.8B-Base`](https://huggingface.co/Qwen/Qwen3.5-0.8B-Base) | BF16 | | [`SketchSSM/Qwen3.5-0.8B-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-0.8B-BF16) | [`Qwen/Qwen3.5-0.8B`](https://huggingface.co/Qwen/Qwen3.5-0.8B) | BF16 | | [`SketchSSM/Qwen3.5-2B-Base-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-2B-Base-BF16) | [`Qwen/Qwen3.5-2B-Base`](https://huggingface.co/Qwen/Qwen3.5-2B-Base) | BF16 | | [`SketchSSM/Qwen3.5-2B-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-2B-BF16) | [`Qwen/Qwen3.5-2B`](https://huggingface.co/Qwen/Qwen3.5-2B) | BF16 | | [`SketchSSM/Qwen3.5-4B-Base-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-4B-Base-BF16) | [`Qwen/Qwen3.5-4B-Base`](https://huggingface.co/Qwen/Qwen3.5-4B-Base) | BF16 | | [`SketchSSM/Qwen3.5-4B-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-4B-BF16) | [`Qwen/Qwen3.5-4B`](https://huggingface.co/Qwen/Qwen3.5-4B) | BF16 | | [`SketchSSM/Qwen3.5-9B-Base-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-9B-Base-BF16) | [`Qwen/Qwen3.5-9B-Base`](https://huggingface.co/Qwen/Qwen3.5-9B-Base) | BF16 | | [`SketchSSM/Qwen3.5-9B-BF16`](https://huggingface.co/SketchSSM/Qwen3.5-9B-BF16) | [`Qwen/Qwen3.5-9B`](https://huggingface.co/Qwen/Qwen3.5-9B) | BF16 | | [`SketchSSM/Qwen3.6-27B-BF16`](https://huggingface.co/SketchSSM/Qwen3.6-27B-BF16) | [`Qwen/Qwen3.6-27B`](https://huggingface.co/Qwen/Qwen3.6-27B) | BF16 | | [`SketchSSM/Qwen3.6-35B-A3B-BF16`](https://huggingface.co/SketchSSM/Qwen3.6-35B-A3B-BF16) | [`Qwen/Qwen3.6-35B-A3B`](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) | BF16 | | [`SketchSSM/Qwen3.8-Flash-Next-NVFP4`](https://huggingface.co/SketchSSM/Qwen3.8-Flash-Next-NVFP4) | [`RadixArk/Qwen3.8-Flash-Next-NVFP4`](https://huggingface.co/RadixArk/Qwen3.8-Flash-Next-NVFP4) | NVFP4 | ## KDA [Collection](https://huggingface.co/collections/SketchSSM/sketchssm-kda-6acc6b2d7b0e3f051d18bee8) | Calibration | Base checkpoint | Weights | | --- | --- | --- | | [`SketchSSM/GLM-5.3-Flash-NVFP4`](https://huggingface.co/SketchSSM/GLM-5.3-Flash-NVFP4) | [`RedHatAI/GLM-5.3-Flash-NVFP4`](https://huggingface.co/RedHatAI/GLM-5.3-Flash-NVFP4) | NVFP4 |