Startup claims its Zeus GPU is 10X faster than Nvidia's RTX 5090: Bolt's first GPU coming in 2026

Bolt Graphics, a GPU startup from Sunnyvale, California, this week introduced its Zeus GPU platform that is designed for gaming, rendering, and supercomputer simulations. The company says that its Zeus GPU not only supports features like upgradeable memory and built-in Ethernet interfaces, but it can also beat Nvidia's GeForce RTX 5090 by around 10 times in path tracing workloads, according to slide published by ServeTheHome.

There is one catch: Zeus can only beat the RTX 5090 GPU in path tracing and FP64 compute workloads. It's not clear how well it will handle traditional rendering techniques, as that was less of a focus. In speaking with Bolt Graphics, the card does support rasterization, but there was less emphasis on that aspect of the GPU, and it may struggle to compete with the best graphics cards when it comes to gaming. And when it comes to data center options like Nvidia's Blackwell B200, it's an entirely different matter.

RISC-V multi-chiplet GPU

Unlike GPUs from AMD, Intel, and Nvidia that rely on proprietary instruction set architectures, Bolt's Zeus relies on the open-source RISC-V ISA, according to the published slides. The Zeus core relies on an open-source out-of-order general-purpose RVA23 scalar core mated with FP64 ALUs and the RVV 1.0 (RISC-V Vector Extension Version 1.0) that can handle 8-bit, 16-bit, 32-bit, and 64-bit data types as well as Bolt's additional proprietary extensions designed for acceleration of scientific workloads.

As Zeus is aimed at path tracing rendering technique as well as compute workloads, we weren't sure whether it has traditional fixed-function GPU hardware like texture units (TMUs) and raster operation units (ROPs). The company reached out to us and said that Zeus does have TMU and ROP engines, but it's not disclosing any of the specifications just yet and is currently working on further optimizations. Each Zeus GPU has one DisplayPort 2.1a and one HDMI 2.1b output.

Like many processors these days, Zeus relies on a multi-chiplet design. The entry-level Zeus 1c26-032 features a single processing unit that is equipped with 32GB of LPDDR5X memory with 273 GB/s of bandwidth, and can be mated with up to 128GB of DDR5 memory using two SO-DIMMs (at 80 GB/s). The Zeus GPU is also paired with an I/O chiplet that packs a QSFP-DD port for 400GbE/800GbE, two PCIe Gen5 x16 slots with CXL 3.0 on top (enabling efficient memory sharing across multiple cards), and a GbE port for BMC. The GPU chiplet connects to its I/O chiplet at 256 GB/s.

The more advanced Zeus 2c26-064/128 uses two Zeus processing units, an I/O chiplet, and supports either 64GB or 128GB of LPDDR5X memory. The most powerful variant — Zeus 4c26-256 — integrates four processing units, four I/O chiplets, 256GB LPDDR5X and up to 2TB of DDR5 memory. The quad-chiplet Zeus implementation is not a card but rather functions as a server.

Unlike high-end GPUs that prioritize bandwidth, Bolt is evidently focusing on greater memory size to handle larger datasets for rendering and simulations. Also, built-in 400GbE and 800GbE ports to enable faster data transfer across networked GPUs indicates the data center focus of Zeus.

Performance

High-quality rendering, real-time path tracing, and compute are key focus areas for Zeus. As a result, even the entry-level Zeus 1c26-32 offers significantly higher FP64 compute performance than Nvidia's GeForce RTX 5090 — up to 5 TFLOPS vs. 1.6 TFLOPS — and considerably higher path tracing performance: 77 Gigarays vs. 32 Gigarays. Zeus also features a larger on-chip cache than Nvidia's flagship — up to 128MB vs. 96MB — and lower power consumption of 120W vs. 575W, making it more efficient for simulations, path tracing, and offline rendering.

However, the RTX 5090 dominates in AI workloads with its 105 FP16 TFLOPS and 1,637 INT8 TFLOPS compared to the 10 FP16 TFLOPS and 614 INT8 TFLOPS offered by a single-chiplet Zeus. The 1c26-32 with with its 10 FP32 TFLOPS performance would have seemingly have no chance against the RTX 5090's 105 TFLOPS.

But we also don't know what sort of traditional graphics performance Zeus might offer. The stated matrix throughput looks relatively tame compared to modern consumer graphics cards, never mind the data center alternatives. A single Blackwell B200 with a 700W power limit offers up 60 TFLOPS of shader FP32 compute, 30 TFLOPS of FP64 dense matrix compute, and 1.8 PetaFLOPS of sparse TF32 compute.

Simply put, there's a lot we don't know yet, and Zeus is only running in simulation right now, with physical hardware slated to arrive later this year. How will it handle traditional rendering, along with ray tracing and AI? We'll need to wait and see, and drivers will be another potential factor — even a large company like Intel tool a long time to work out a lot of the driver kinks.

The quad-chiplet 4c26-256 — which still consumes less power than the GeForce RTX 5090, but is likely to be more expensive — on paper can beat Nvidia's gaming flagship in all workloads except FP32 and FP16, which emphasizes that traditional gaming was not the priority. The company indicates that the quad-chiplet variant is optimized for electromagnetic field modeling, photonics research, and FFT calculations. With its larger memory pools and reduced reliance on external storage, Zeus promises to improve speed of large-scale simulations. Assuming, of course, that its hybrid memory subsystem works efficiently across all workloads.

Software

Software support will be a critical factor in the success of Bolt's Zeus GPUs, as hardware capabilities alone are not enough to compete with established players like AMD and Nvidia. Unlike CUDA for Nvidia and ROCm for AMD, Bolt's Zeus lacks a mature, widely adopted software ecosystem. Since it is based on RISC-V, Zeus can potentially leverage existing open-source tools and libraries, but without strong developer support, adoption will be limited.

One thing that Bolt introduced is its Glowstick path tracing rendering engine, which could be a promising in-house solution for real-time rendering. Traditional rendering workflows require long processing times to visualize changes, whereas Zeus significantly reduces this delay, which makes it useful for professional visualization applications. Compared to existing solutions, Bolt claims up to 2.5x faster performance on single-chip variants, scaling higher with multiple GPUs.

However, it is unclear whether Zeus can support industry-standard frameworks such as OpenCL, Vulkan, and CUDA-translation layers, which are essential for gaining traction in HPC workloads. If Bolt Graphics provides strong developer tools, robust compiler support, and compatibility with Linux-based HPC environments, Zeus could establish itself as a competitor in scientific computing and rendering, but it faces an uphill battle against Nvidia's well-established ecosystem.

Availability

Bolt Graphics says that the first developer kits will be available in late 2025, with full production set for late 2026, which will give time for software developers to play with the hardware. Right now, Zeus is only running in simulation, so none of the performance claims have been verified with actual hardware.

The number of discrete GPU developers from the U.S. and Western Europe shrank to three companies in 2025 from around 10 in 2000. Nvidia dominates the market and is followed by AMD and Intel, which can compete in select segments only. No company in the recent years — at least outside of China — was bold enough to engage into competition against these three contenders, so the very emergence of Bolt Graphics seems like a breakthrough. However, the major focuses of Bolt's Zeus are high-quality rendering for movie and scientific industries as well as high-performance supercomputer simulations.

If Zeus delivers on its promises, it could establish itself as a serious alternative for scientific computing, path tracing, and offline rendering. But without strong software support, it risks struggling against dominant market leaders.