rob thijssen
66632db578
All checks were successful
build-prerelease / Resolve version stamps + change detection (push) Successful in 12s
build-prerelease / Build cortex binary (push) Has been skipped
build-prerelease / Package cortex RPM (push) Has been skipped
build-prerelease / Build helexa-bench binary (push) Has been skipped
build-prerelease / Package helexa-bench RPM (push) Has been skipped
build-prerelease / Build neuron-ampere (push) Successful in 2m25s
build-prerelease / Lint (fmt + clippy) (push) Successful in 2m29s
build-prerelease / Test (push) Successful in 6m6s
build-prerelease / Build neuron-blackwell (push) Successful in 9m8s
build-prerelease / Build helexa-upstream binary (push) Has been skipped
build-prerelease / Package helexa-upstream RPM (push) Has been skipped
build-prerelease / Build neuron-ada (push) Successful in 1m56s
build-prerelease / Package helexa-neuron-ada RPM (push) Successful in 44s
build-prerelease / Package helexa-neuron-ampere RPM (push) Successful in 51s
build-prerelease / Package helexa-neuron-blackwell RPM (push) Successful in 2m52s
build-prerelease / Publish to rpm.lair.cafe (unstable) (push) Successful in 26s
Activates the #98 engine for beast's TP models (27B, Coder-Next): admission bound and engine slot count in one knob. Reverting this section (or NEURON_BATCHING=0) restores batch-1 serialization. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TczcGF7JSjJs8r15RSSGpx
33 lines
1001 B
TOML
33 lines
1001 B
TOML
# neuron.toml for beast.hanzalova.internal
|
|
#
|
|
# 2x RTX 5090 (32 GB each) — TP-2 capable. Pre-warms Qwen3.6-27B with
|
|
# q5k ISQ across both GPUs at activation, matching the validate-neuron
|
|
# invocation: `validate-neuron.sh beast.hanzalova.internal
|
|
# Qwen/Qwen3.6-27B q5k 2`.
|
|
#
|
|
# Synced to /etc/neuron/neuron.toml by script/infra-setup.sh. Edits
|
|
# take effect after the next deploy workflow run restarts the service
|
|
# (default_models is read at activation).
|
|
|
|
port = 13131
|
|
|
|
[[harnesses]]
|
|
name = "candle"
|
|
|
|
[harness.candle]
|
|
|
|
# Batched decode engine (#98): up to 8 concurrent text streams
|
|
# multiplex through one lockstep (B,1) forward per decode step.
|
|
# max_in_flight is both the admission bound and the engine's slot
|
|
# count. NEURON_BATCHING=0 (systemd drop-in) is the kill switch;
|
|
# removing this section reverts to batch-1 serialization.
|
|
[harness.candle.admission]
|
|
max_in_flight = 8
|
|
|
|
[[default_models]]
|
|
model_id = "Qwen/Qwen3.6-27B"
|
|
harness = "candle"
|
|
quant = "q6k"
|
|
tensor_parallel = 2
|
|
devices = [0, 1]
|