MLIR
Shares rank 6 of 10 compiler infrastructures by recommenders
Widely used LLVM-based compiler infrastructure for building IRs and DSLs; people praise its power, though critics say it adds heavy complexity and requires C++ and TableGen.
7 people recommended MLIR in Hacker News comments, 3 had mixed feelings and 3 criticized it, November 2025 to October 2026.
Summary written by Claude from 13 comments. It was written before the November 2026 recount, and 12 of those comments still count.
13 people gave an opinion
- recommend it
- 7
- mixed feelings
- 3
- warn against or regret it
- 3
The record
Recommended in 6 of 12 months
In 7 different threads.
- November 2025: not recommended
- December 2025: not recommended
- January 2026: not recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: not recommended
- October 2026: recommended
Why it doesn’t carry the badge
- 7 people recommend itNeeds 10 or more, not met.
- 2.3 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 6 of 12 monthsNeeds 4 or more, met.
- Recommended in 7 threadsNeeds 5 or more, met.
- The biggest thread supplies 14% 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 12 of the 13 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
Best-in-class compiler infrastructure
“MLIR is just genuinely very good at some of this stuff compared to basically anyone else”
Comment from 18 May 2026 in “Jank now has its own custom IR” “there's nothing that comes close to LLVM and MLIR”
Comment from 19 Feb 2026 in “Ladybird: Closing this as we are no longer pursuing Swift adoption” “it's nothing short of magical, in my opinion!”
Comment from 19 Oct 2026 in “MLIR CodeGen Dialects for Machine Learning Compilers (2022)” Good base for IRs and DSLs
“The benefit is that you focus on designing your IR without also building supporting infrastructure”
Comment from 20 Jul 2026 in “UnifiedIR for Julia” “has proven quite valuable to create Python JIT DSL”
Comment from 8 May 2026 in “Mojo 1.0 Beta”
What people warn about
Each mark is one comment
Adds significant complexity overall
“MLIR pulls in a lot of interface complexity that we don't want or need”
Comment from 19 Jul 2026 in “UnifiedIR for Julia” “MLIR added a great deal of complexity on top of designing/implementing an IR”
Comment from 18 May 2026 in “Jank now has its own custom IR” C++ and TableGen hard to work with
“writing Julia is a lot nicer than writing C++”
Comment from 19 Jul 2026 in “UnifiedIR for Julia”
Other compiler infrastructures people recommend
The four most recommended other compiler infrastructures, 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 |
|---|---|---|---|---|---|---|---|
| LLVM | 72 | 34 | 86 | 46% | 11 | — | — |
| QBE | 12 | 4 | 9 | 57% | 9 | — | — |
| Roslyn | 11 | 0 | 1 | 92% | 8 | — | Approved |
| EDG C++ front-end | 11 | 4 | 3 | 79% | 2 | — | — |
| MLIRThis page | 7 | 3 | 3 | 70% | 6 | — | — |
What people said
12 of the 13 opinions about it, picked in proportion to the three counts, newest first. Each links to the original comment.
“it's nothing short of magical, in my opinion!”
“so prevalent in the ML compile stack everywhere”
“using multiple languages is cumbersome and annoying”
“The benefit is that you focus on designing your IR without also building supporting infrastructure”
“writing Julia is a lot nicer than writing C++”
“MLIR pulls in a lot of interface complexity that we don't want or need”
“Chris's previous projects like LLVM and MLIR are fortunate enough to grow and thrive organically”
“MLIR added a great deal of complexity on top of designing/implementing an IR”
“MLIR is just genuinely very good at some of this stuff compared to basically anyone else”
“There's a recent NVIDIA MLIR that is quite good and fast.”
“has proven quite valuable to create Python JIT DSL”
“there's nothing that comes close to LLVM and MLIR”
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.