tt-awesome — Articles & ResourcesArticles, papers, lessons, talks, videos, and demos from the Tenstorrent ecosystemhttps://tenstorrent.github.io/tt-awesome/feeds/articles.xmlTenstorrent Communityhttps://tenstorrent.github.io/tt-awesome/2026-06-03T00:00:00Zhttps://docs.tenstorrent.com/tt-vscode-toolkit/lessons/animatediff-video-generation/tt-animatediff — Native Video Animation with AnimateDiff (VSCode Toolkit)2026-06-03T00:00:00ZGenerates short, temporally coherent animated GIFs using the AnimateDiff model on Tenstorrent hardware. Phase 1 runs the correct SD 1.4 + MotionAdapter architecture on CPU; Phase 2 accelerates spatial denoising on Blackhole using the TTNN UNet. Produces vibrant 8-frame animations in ~15 s/frame on a P300C.https://anuraagw.me/blog/blackhole-architectureTenstorrent Blackhole Architecture Guide — anuraagw.me — February 20262026-05-13T00:00:00ZA 6,500-word community deep dive into the Blackhole p100a architecture: the tile model (Tensix, DRAM, SiFive x280 L2CPU, Ethernet, PCIe, NoC arc), firmware startup sequence, MOP micro-op processor, replay buffer, FPU/SFPU sync, and the anatomy of a kernel. From the author of blackhole-py.https://www.youtube.com/watch?v=CixEFPc8oxgTenstorrent Architecture — W&M CSCI654 Advanced Computer Architecture — Lecture 20 — Tenstorrent Architecture (YouTube)2026-05-13T00:00:00ZLecture 20 from William & Mary's graduate Computer Architecture course. Frames Tenstorrent in the landscape between GPUs and TPUs, draws comparisons to Cerebras and SambaNova, then dives deep into the Wormhole chip and Tensix core: the 5 RISC-V core design, SFPU, NoC, and dataflow execution model.https://arxiv.org/abs/2407.13885Attention in SRAM on Tenstorrent Grayskull — arXiv:2407.138852026-05-13T00:00:00ZA fused kernel for the Grayskull architecture implementing Transformer self-attention entirely within SRAM. Combines matrix multiply, attention score scaling, and Softmax without DRAM accesses, achieving significant speedups over non-fused implementations.https://arxiv.org/abs/2505.06085Assessing Tenstorrent Grayskull RISC-V MatMul Acceleration for LLMs — arXiv:2505.060852026-05-13T00:00:00ZEvaluates the Tenstorrent Grayskull e75 RISC-V accelerator for matrix multiplication at reduced numerical precision (BFP8 and LoFi), a fundamental kernel in LLM inference computation.https://arxiv.org/abs/2605.02744Porting Strategies for Gravitational N-Body Simulations on Tenstorrent Wormhole — arXiv:2605.027442026-05-13T00:00:00ZEvaluates three strategies for scaling an N-body code across multiple Tenstorrent Wormhole accelerators. Builds on the established performance of single-card N-body work to explore parallelism via the on-chip NoC and multi-accelerator configurations.https://arxiv.org/abs/2512.22168TileLoom: Automatic Dataflow Planning for Spatial Dataflow Accelerators — arXiv:2512.221682026-05-13T00:00:00ZCompiler system that automatically generates efficient dataflow plans for tile-based languages on spatial accelerators including Tenstorrent Wormhole. Exploits on-chip network forwarding between processing elements to reduce DRAM pressure.https://arxiv.org/abs/2604.03279Rewriting TTS Inference Economics: Lightning V2 on Tenstorrent vs. NVIDIA L40S — arXiv:2604.032792026-05-13T00:00:00ZShows that Text-to-Speech inference on Tenstorrent Lightning V2 achieves 4× lower cost than NVIDIA L40S. Applies BlockFloat8 (BFP8) and low-fidelity (LoFi) precision strategies to TTS despite their greater numerical fragility compared to LLMs.https://github.com/tenstorrent/tt-llk/blob/main/docs/llk/l2/top_level_overview.mdtt-llk — Top-level architecture overview2026-05-13T00:00:00ZTenstorrent Low-Level Kernels: the C++ library that directly programs the RISC-V cores inside each Tensix compute engine. TRISC0 (unpack), TRISC1 (math/FPU/SFPU), and TRISC2 (pack) are all programmed through this layer — it is the interface between TT-Metal kernel code and bare silicon.https://arxiv.org/abs/2605.07599Stencil Computations on Tenstorrent Wormhole — arXiv:2605.075992026-05-12T00:00:00ZMaps 2D 5-point stencil computations onto the Tenstorrent Wormhole RISC-V AI dataflow accelerator via two implementations: element-wise decomposition (Axpy) and matrix-multiplication reformulation (MatMul). Profiling shows the isolated Wormhole kernel is competitive with CPU execution, with PCIe transfers and initialization driving end-to-end overhead; Axpy achieves lower energy than the CPU baseline at large scales. Identifies architectural and software directions for making AI accelerators viable for HPC stencil workloads. 2025.https://fosdem.org/2026/schedule/event/AJLNVH-tt-boltz/tt-boltz — FOSDEM 2026 — Drug Discovery on Tenstorrent Hardware2026-05-08T00:00:00ZBoltz-2 biomolecular model for drug discovery on Tenstorrent Blackhole. Supports single-card and multi-card configurations — QuietBox (4×) and Galaxy (32×). Approaches physics-based FEP accuracy at 1000× the speed.https://clehaxze.tw/gemlog/2025/04-21-programming-tensotrrent-processors.gmiProgramming Tenstorrent Processors — clehaxze.tw — April 20252026-05-08T00:00:00ZDeep-dive into the Tenstorrent architecture and Metalium programming model — circular buffers, kernel synchronization, NoC routing, and where the footguns are. The honest guide to thinking in Tensix.https://www.jasondavies.com/2025/tenstorrent-where/Tenstorrent SFPU Kernel Series — Jason Davies — Optimal "where" on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://www.jasondavies.com/2025/tenstorrent-multiply-int32/Tenstorrent SFPU Kernel Series — Jason Davies — 32-bit Integer Multiplication on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://www.jasondavies.com/2025/tenstorrent-typecast/Tenstorrent SFPU Kernel Series — Jason Davies — Typecast on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://www.jasondavies.com/2026/tenstorrent-multiply-int16/Tenstorrent SFPU Kernel Series — Jason Davies — 16-bit Integer Multiplication on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://www.jasondavies.com/2026/tenstorrent-cbrt/Tenstorrent SFPU Kernel Series — Jason Davies — Cube Root on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://www.jasondavies.com/2026/tenstorrent-sin-cos-tan/Tenstorrent SFPU Kernel Series — Jason Davies — Accurate sin/cos/tan on Tenstorrent2026-05-08T00:00:00ZSponsored series of deep technical articles on implementing optimal SFPU kernels for the Tenstorrent Wormhole and Blackhole vector units. Covers where, typecasting, 16/32-bit integer multiplication, cube root, and accurate sin/cos/tan — with cycle counts, assembly walkthroughs, and Blackhole vs Wormhole comparisons throughout.https://clehaxze.tw/gemlog/2024/07-07-a-gentle-guide-on-getting-your-tenstorrent-card-running-on-arch-linux-with-the-metalium-stack.gmiA Gentle Guide: Tenstorrent Card on Arch Linux with Metalium — clehaxze.tw — July 20242026-05-08T00:00:00ZStep-by-step guide to getting a Tenstorrent card running on Arch Linux with the full Metalium stack. Practical troubleshooting from someone who did it the hard way first.https://clehaxze.tw/gemlog/2024/06-02-thoughts-and-logs-after-messing-with-tenstorrent-grayskull.gmiThoughts and Logs After Messing with Tenstorrent Grayskull — clehaxze.tw — June 20242026-05-08T00:00:00ZHonest field notes from getting a Grayskull card running and writing first Metalium kernels. Covers setup pitfalls, processor hangs, memory protection quirks, and what makes Metalium compelling despite early rough edges.https://arxiv.org/abs/2506.15437Exploring Fast Fourier Transforms on the Tenstorrent Wormhole — arXiv:2506.154372026-05-08T00:00:00ZPorts the Cooley-Tukey FFT algorithm to the Wormhole n300 RISC-V accelerator. The Wormhole draws 8× less power and consumes 2.8× less energy than a 24-core Xeon Platinum for a 2D FFT. ISC 2025.https://www.research.ed.ac.uk/en/publications/exploring-fast-fourier-transforms-on-the-tenstorrent-wormhole/Exploring Fast Fourier Transforms on the Tenstorrent Wormhole — University of Edinburgh2026-05-08T00:00:00ZPorts the Cooley-Tukey FFT algorithm to the Wormhole n300 RISC-V accelerator. The Wormhole draws 8× less power and consumes 2.8× less energy than a 24-core Xeon Platinum for a 2D FFT. ISC 2025.https://arxiv.org/abs/2509.19294Accelerating Gravitational N-Body Simulations on Tenstorrent Wormhole — arXiv:2509.192942026-05-08T00:00:00ZAccelerates an astrophysical N-body simulation on the Wormhole n300. Achieves 2× speedup and 2× energy savings over a highly optimized CPU implementation. SC '25 Workshop.https://dl.acm.org/doi/10.1145/3731599.3767528Accelerating Gravitational N-Body Simulations on Tenstorrent Wormhole — ACM SC '252026-05-08T00:00:00ZAccelerates an astrophysical N-body simulation on the Wormhole n300. Achieves 2× speedup and 2× energy savings over a highly optimized CPU implementation. SC '25 Workshop.https://arxiv.org/abs/2603.23343Numerical Kernels on a Spatial Accelerator: Tenstorrent Wormhole — arXiv:2603.233432026-05-08T00:00:00ZImplements three numerical kernels and composes them into a conjugate gradient solver on Wormhole. Demonstrates AI accelerators merit consideration for HPC workloads traditionally dominated by CPUs and GPUs. 2026.https://arxiv.org/abs/2409.18835Accelerating Stencils on the Tenstorrent Grayskull RISC-V Accelerator — arXiv:2409.188352026-05-08T00:00:00ZExplores stencil computation on the Grayskull PCIe RISC-V accelerator. Early academic work examining TT hardware for HPC stencil workloads. 2024.https://dl.acm.org/doi/10.1145/3805621.3807614SwiftNPU: Scalable Shape-Flexible Allocation for Inter-Core Connected NPUs — ACM DL2026-05-08T00:00:00ZMakes multi-tenant NPU sharing practical for Blackhole-class hardware using polynomial-time allocation algorithms. Delivers up to 1.37× higher utilization and 1.14× faster workload completion. Up to 890,000× faster than NP-hard baselines.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/qb2-local-agents/Local AI Agents on Tenstorrent — Local AI Agents on QuietBox 22026-05-08T00:00:00ZThree agentic projects running fully on-device: local AI agents on QuietBox 2, a coding assistant powered by Aider against a local inference server, and the OpenClaw AI assistant on QuietBox 2. No cloud APIs — all inference runs on TT hardware.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/coding-assistant/Local AI Agents on Tenstorrent — Coding Assistant with Aider2026-05-08T00:00:00ZThree agentic projects running fully on-device: local AI agents on QuietBox 2, a coding assistant powered by Aider against a local inference server, and the OpenClaw AI assistant on QuietBox 2. No cloud APIs — all inference runs on TT hardware.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/video-generation-ttmetal/Video Generation on Tenstorrent — Video Generation via Frame-by-Frame Diffusion2026-05-08T00:00:00ZThree lesson-projects covering on-device video synthesis: frame-by-frame diffusion with tt-local-generator, native AnimateDiff video animation, and video generation on QuietBox 2. All run entirely on TT hardware with no cloud dependency.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/qb2-video-generation/Video Generation on Tenstorrent — Video Generation on QuietBox 22026-05-08T00:00:00ZThree lesson-projects covering on-device video synthesis: frame-by-frame diffusion with tt-local-generator, native AnimateDiff video animation, and video generation on QuietBox 2. All run entirely on TT hardware with no cloud dependency.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cookbook-particle-life/Tenstorrent Cookbook: Particle Life Simulator — Cookbook Recipe 5: Particle Life Simulator2026-05-08T00:00:00ZParticle Life simulation on Tenstorrent hardware — an emergent-behavior N-body system where simple attraction/repulsion rules between species produce complex lifelike patterns. Cookbook recipe demonstrating parallel N-body compute on Tensix.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-01-computer/CS Fundamentals on Tenstorrent Hardware — Module 1: RISC-V & Computer Architecture2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-02-memory/CS Fundamentals on Tenstorrent Hardware — Module 2: The Memory Hierarchy2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-03-parallelism/CS Fundamentals on Tenstorrent Hardware — Module 3: Parallel Computing2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-04-networks/CS Fundamentals on Tenstorrent Hardware — Module 4: Networks and Communication2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-05-synchronization/CS Fundamentals on Tenstorrent Hardware — Module 5: Synchronization2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-06-abstraction/CS Fundamentals on Tenstorrent Hardware — Module 6: Abstraction Layers2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cs-fundamentals-07-complexity/CS Fundamentals on Tenstorrent Hardware — Module 7: Computational Complexity in Practice2026-05-08T00:00:00ZSeven-module computer science curriculum taught on real Tenstorrent hardware. Covers RISC-V architecture, memory hierarchy, parallel computing, networks and NoC, synchronization, abstraction layers, and computational complexity — all grounded in what is physically happening on the chip.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/qb2-openclaw-assistant/tt-claw — OpenClaw AI Assistant on QuietBox 22026-05-08T00:00:00ZA Tenstorrent-powered claw machine that rewards players with real prizes. The QuietBox 2 runs local AI inference to act as an agent controlling the claw hardware — the OpenClaw AI assistant lesson builds directly on this project.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/image-generation/Stable Diffusion XL on Tenstorrent — Image Generation with Stable Diffusion XL2026-05-08T00:00:00ZOn-device image generation with Stable Diffusion XL running entirely on Tenstorrent hardware. Full inference pipeline with no cloud dependency.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/forge-image-classification/Image Classification with TT-Forge — Image Classification with TT-Forge2026-05-08T00:00:00ZEnd-to-end image classification project using TT-Forge — compile and run a PyTorch classification model on Tenstorrent hardware with no kernel authoring required.https://docs.tenstorrent.com/tt-vscode-toolkit/tensix-playground/Tensix Grid Playground — Tensix Grid Playground (interactive)2026-05-08T00:00:00ZInteractive browser-based visualizer of the Tenstorrent Tensix grid architecture. Explore the NoC, core layout, and dataflow patterns without hardware — a great companion for learning kernel programming.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/cookbook-game-of-life/Tenstorrent Cookbook: Conway's Game of Life — Cookbook Recipe 1: Conway's Game of Life2026-05-08T00:00:00ZTT-Metalium implementation of Conway's Game of Life as a cookbook recipe. Each generation is a full parallel kernel dispatch over the grid — a clean introduction to stateful compute on Tensix cores.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct1-understanding-training/Custom Model Training on Tenstorrent — Understanding Custom Training2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct2-dataset-fundamentals/Custom Model Training on Tenstorrent — Dataset Fundamentals2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct3-configuration-patterns/Custom Model Training on Tenstorrent — Configuration Patterns2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct4-finetuning-basics/Custom Model Training on Tenstorrent — Fine-tuning Basics2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct5-multi-device-training/Custom Model Training on Tenstorrent — Multi-Device Training2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.https://docs.tenstorrent.com/tt-vscode-toolkit/lessons/ct6-experiment-tracking/Custom Model Training on Tenstorrent — Experiment Tracking2026-05-08T00:00:00ZEight-lesson series covering the full custom training workflow on TT hardware: dataset fundamentals, configuration patterns, fine-tuning, multi-device distributed training, experiment tracking, model architecture basics, and training from scratch.