---
format: "aidr-story-markdown/v1"
id: "85cbd2dd2c894a083bf84d67c62b144ceea30f64ae305e2ab16da31a8e5c6e54"
canonical_url: "https://aidr.today/85cbd2dd?lang=en"
title: "Open-Source Projects Unlock AI Performance on Nvidia RTX 20, 30, and 40 GPUs"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-25T17:23:44.000Z"
category: "Open Source"
topics: ["nvidia","open-source"]
source_urls: ["https://marketbrief.now/tech/open-source-projects-unlock-ai-performance-on-nvidia-rtx-20-30-and-40-gp-505eae13","https://x.com/lxfater/status/2113891680000901380","https://x.com/MaxForAI/status/2114402686821159108","https://x.com/yacineMTB/status/2114353625849893315","https://huggingnews.com/tech/open-source-projects-unlock-ai-performance-on-nvidia-rtx-20-30-and-40-gp-505eae13"]
summary: "Open-source projects have expanded a performance exploit that removes AI computation caps on Nvidia's RTX 20, 30, and 40 series graphics cards by bypassing Tensor Core hardware limits without flashing the BIOS or modifying official drivers. The unlock enables some models, including the RTX 2080 Ti, to nearly double their FP16 operation throughput, with the RTX 4090's compute performance rising from 165.6 to 331.6 TFLOPS and the RTX 3090 Ti's BF16 performance increasing from 87.5 to 157.4 TFLOPS. Previously limited to Linux environments requiring the disabling of Secure Boot, the projects have since been reproduced on Windows 11."
---

# Open\-Source Projects Unlock AI Performance on Nvidia RTX 20, 30, and 40 GPUs

> [Open the canonical story](<https://aidr.today/85cbd2dd?lang=en>)

**Published:** 2026-10-25T17:23:44.000Z
**Category:** Open Source
**Topics:** nvidia, open\-source

## Summary

Open\-source projects have expanded a performance exploit that removes AI computation caps on Nvidia's RTX 20, 30, and 40 series graphics cards by bypassing Tensor Core hardware limits without flashing the BIOS or modifying official drivers\. The unlock enables some models, including the RTX 2080 Ti, to nearly double their FP16 operation throughput, with the RTX 4090's compute performance rising from 165\.6 to 331\.6 TFLOPS and the RTX 3090 Ti's BF16 performance increasing from 87\.5 to 157\.4 TFLOPS\. Previously limited to Linux environments requiring the disabling of Secure Boot, the projects have since been reproduced on Windows 11\.

## Sources

- [Story source](<https://marketbrief.now/tech/open-source-projects-unlock-ai-performance-on-nvidia-rtx-20-30-and-40-gp-505eae13>)
- [Supporting source](<https://x.com/lxfater/status/2113891680000901380>)
- [Story source](<https://x.com/MaxForAI/status/2114402686821159108>)
- [Supporting source](<https://x.com/yacineMTB/status/2114353625849893315>)
- [Story source](<https://huggingnews.com/tech/open-source-projects-unlock-ai-performance-on-nvidia-rtx-20-30-and-40-gp-505eae13>)