ZLUDA
Ranked 5 of 9 GPU compute frameworks by recommenders
Recommended as the practical open-source drop-in for running CUDA on AMD GPUs, though critics note incomplete coverage and no CDNA support.
10 people recommended ZLUDA in Hacker News comments, 3 had mixed feelings and 6 criticized it, November 2025 to October 2026.
Summary written by Claude from 19 comments.
19 people gave an opinion
- recommend it
- 10
- mixed feelings
- 3
- warn against or regret it
- 6
The record
Recommended in 6 of 12 months
In 7 different threads.
- November 2025: recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: recommended
- October 2026: recommended
Why it doesn’t carry the badge
- 10 people recommend itNeeds 10 or more, met.
- 1.6 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 6 of 12 monthsNeeds 4 or more, met.
- Recommended in 7 threadsNeeds 5 or more, met.
- The biggest thread supplies 30% of its recommendersNeeds half or less, met.
The same rules apply to every product, and the badge can’t be bought. Mixed feelings count toward neither side. How the badge works
What people say about it
Reasons grouped from a sample of 19 of the 21 opinions about it, picked in proportion to how many people recommend it, have mixed feelings and criticize it. Each mark is one comment in the sample. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Practical drop-in CUDA replacement
“ZLUDA is the “right” drop in CUDA replacement for almost any practical use case”
Comment from 18 Feb 2026 in “BarraCUDA Open-source CUDA compiler targeting AMD GPUs” “if you want a practical path towards drop-in replacing CUDA, you want ZLUDA”
Comment from 17 Feb 2026 in “BarraCUDA Open-source CUDA compiler targeting AMD GPUs” “this is closest thing we have to "cuda on non-nvidia" hardware”
Comment from 14 Jul 2026 in “Alternative(s) to run CUDA on non-Nvidia hardware” - Comment from 18 Feb 2026 in “BarraCUDA Open-source CUDA compiler targeting AMD GPUs”
Open source, works with binaries
“zluda is open source with extra features like game physics”
Comment from 26 Jul 2026 in “DeepSeek pause fundraise after comments on compute gap to US leaked (transcript) [pdf]” “which is open source and works on pre-compiled binaries”
Comment from 14 Jul 2026 in “Alternative(s) to run CUDA on non-Nvidia hardware”
What people warn about
Each mark is one comment
Incomplete CUDA API coverage
“ZLUDA doesn't have full coverage though and that means only a subset of cuda codebases can be ported successfully”
Comment from 18 Feb 2026 in “BarraCUDA Open-source CUDA compiler targeting AMD GPUs” “ZLUDA basically crash on high memory demand, and only accounted for ~70% of CUDA API coverage”
Comment from 24 Jul 2026 in “AMD's Instinct MI455X: Aiming for the Sun” Earlier attempts failed
“most previous attempts (like ZLUDA) hit a wall”
Comment from 19 Feb 2026 in “BarraCUDA Open-source CUDA compiler targeting AMD GPUs” “Everyone who tried to compete failed hard”
Comment from 5 Nov 2025 in “Michael Burry a.k.a. "Big Short",discloses $1.1B bet against Nvidia&Palantir”
Other GPU compute frameworks people recommend
The four most recommended other GPU compute frameworks, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Mixed | Criticize | Approval | Months | Had for | Badge |
|---|---|---|---|---|---|---|---|
| NVIDIA CUDA | 65 | 22 | 69 | 49% | 12 | — | — |
| AMD ROCm | 26 | 17 | 86 | 23% | 8 | — | — |
| BarraCUDA | 15 | 5 | 4 | 79% | 2 | — | — |
| Triton | 13 | 5 | 2 | 87% | 8 | — | Approved |
| ZLUDAThis page | 10 | 3 | 6 | 63% | 6 | — | — |
What people said
19 of the 21 opinions about it, picked in proportion to the three counts, newest first. Each links to the original comment.
“seems to work at near-native performance from what I can find on Google”
“All of the buttons and links on that page redirect to spam pages.”
“it doesn't support CDNA, which is a showstopper for enterprises”
“it’s been making steady progress”
“zluda is open source with extra features like game physics”
“ZLUDA basically crash on high memory demand, and only accounted for ~70% of CUDA API coverage”
“this is closest thing we have to "cuda on non-nvidia" hardware”
“which is open source and works on pre-compiled binaries”
“the real value of Zluda to the community is going to lie on more niche applications”
“most previous attempts (like ZLUDA) hit a wall”
“Translating CUDA to AMD hardware would only advertise why Nvidia is worth so much.”
“ZLUDA doesn't have full coverage though and that means only a subset of cuda codebases can be ported successfully”
“Was really inspired by ZLUDA when I made this.”
“I got Zluda to generally work with comfyui well enough.”
“ZLUDA is the “right” drop in CUDA replacement for almost any practical use case”
“Agreed on ZLUDA being the practical choice”
“if you want a practical path towards drop-in replacing CUDA, you want ZLUDA”
“Just put in some work to polish what already exists”
“Everyone who tried to compete failed hard”
Counts come from 1,791,911 Hacker News comments labeled by Claude, November 2025 to October 2026, with data through October 31, 2026. Counts updated 2 Nov 2026. How we count. Report a mistake.