kaio-candle
v0.2.0 ExperimentalCandle bridge for KAIO — CustomOp bindings for 12 GPU ops (matmul_tc, matmul_tc_bf16, matmul_tc_async, matmul_tc_bf16_async, matmul_int4, matmul_int8, attention_tc, attention_tc_causal, attention_flash, attention_flash_causal, qkv_project_int8, qkv_project_int4). All four matmul TC variants (f16+bf16, sync+async) support backward (autograd); FlashAttention (plain+causal) supports backward via dedicated PTX kernels. Build with `cargo build --features cuda`.
Quick Verdict
- !Pre-1.0: API may have breaking changes
- ✓Permissive license (MIT OR Apache-2.0)
Security
Deep Insights
6 downloads in the last 30 days (0/day), up 100% from the previous period.
At 41KB, kaio-candle is lightweight. Small crate size correlates with focused, well-scoped functionality.
Health Breakdown
Recency, release consistency, active ratio
Yanked ratio, deps, size, maturity, features
Reverse deps, ownership, ecosystem
Downloads, momentum, growth trend
Docs, repo, license, metadata