rob thijssen
10c151efa5
Some checks failed
build-prerelease / Package cortex RPM (push) Blocked by required conditions
build-prerelease / Resolve version stamps (push) Successful in 36s
CI / Format (push) Successful in 39s
CI / Clippy (push) Successful in 2m21s
build-prerelease / Build neuron-blackwell (push) Successful in 3m36s
CI / Test (push) Successful in 4m39s
CI / Build cortex SRPM (push) Has been skipped
CI / Build neuron SRPM (push) Has been skipped
CI / Publish cortex to COPR (push) Has been skipped
CI / Publish neuron to COPR (push) Has been skipped
CI / Bump version in source (push) Has been skipped
build-prerelease / Build cortex binary (push) Successful in 4m34s
build-prerelease / Build neuron-ada (push) Has been cancelled
build-prerelease / Package helexa-neuron-ada RPM (push) Has been cancelled
build-prerelease / Package helexa-neuron-ampere RPM (push) Has been cancelled
build-prerelease / Package helexa-neuron-blackwell RPM (push) Has been cancelled
build-prerelease / Publish to rpm.lair.cafe (unstable) (push) Has been cancelled
build-prerelease / Build neuron-ampere (push) Has been cancelled
TP per-token Rust loop replaced with shared run_delta_rule dispatch from arch/qwen3_5/linear_attn.rs. Both single-GPU and TP variants now use the cuda kernel when available, per-token Rust fallback otherwise. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>