Search results

FP8 still works in pytorch 2.9.1? I get an error: llvm.fmul' op operand #0 must be floating point LLVM type or LLVM dialect-compatible vector of floating point LLVM type, but got 'i8'

Status: Open.

#176 In woct0rdho/triton-windows;

· Ph0rk0z opened on Dec 14, 2025

this shouldn't be compiled against a vcredist that requires copying and pasting dlls into the python installation

bugSomething isn't workingSomething isn't working

Status: Open.

#174 In woct0rdho/triton-windows;

· doctorpangloss opened on Dec 3, 2025

FP8 causes more recompilation than GGUF?

Status: Open.

#168 In woct0rdho/triton-windows;

· muljanis45 opened on Nov 4, 2025

fatal error LNK1181 test_triton.py

Status: Open.

#164 In woct0rdho/triton-windows;

· Rahimmess opened on Oct 29, 2025

Possible Pascal GPU support

Status: Open.

#133 In woct0rdho/triton-windows;

· xeghesoopn opened on Jun 27, 2025

Provide wheel for Windows ARM64

Status: Open.

#119 In woct0rdho/triton-windows;

· khmyznikov opened on May 7, 2025