Job instance 582551
| Name | wu_10_g492562_90128_0 |
| Workunit | 593786 |
| Created | 7 Feb 2026, 8:43:15 UTC |
| Sent | 7 Feb 2026, 8:43:19 UTC |
| Report deadline | 14 Feb 2026, 8:43:19 UTC |
| Received | 7 Feb 2026, 8:47:44 UTC |
| Server state | Over |
| Outcome | Success |
| Client state | Done |
| Exit status | 0 (0x00000000) |
| Computer ID | 34 |
| Run time | 3 min 21 sec |
| CPU time | 2 min 48 sec |
| Priority | 0 |
| Validate state | Valid |
| Credit | 25.44 |
| Device peak FLOPS | 1,129.60 GFLOPS |
| Application version | Axiom Federated AI (GPU) v3.41 (cuda) x86_64-pc-linux-gnu |
| Peak working set size | 1.33 GB |
| Peak swap size | 12.55 GB |
| Peak disk usage | 17.13 KB |
Stderr output
<core_client_version>8.2.8</core_client_version> <![CDATA[ <stderr_txt> 2026-02-07 08:44:15 (3076402): BOINC streaming wrapper starting 2026-02-07 08:44:15 (3076402): Streaming wrapper: running ../../projects/axiom.heliex.net/axiom_worker_3.41_cuda_x86_64-pc-linux-gnu (wu.json result.json) 2026-02-07 08:44:15 (3076402): Streaming wrapper: created child process 3076404 [GPU] CuPy CUDA: 1999MB total, 1951MB free [BOINC] Starting BOINC mode [BOINC] GPU detected: 1999MB VRAM, 1951MB free [BOINC] Work unit: wu.json [BOINC] Result file: result.json [BOINC] Current directory: /var/lib/boinc/slots/8 [BOINC] Directory contents: ['axiom_worker_3.41_cuda_x86_64-pc-linux-gnu', 'boinc_lockfile', 'boinc_setup_complete', 'job.xml', 'streaming_wrapper_300_x86_64-pc-linux-gnu', 'init_data.xml', 'boinc_mmap_file', 'stderr.txt', 'result.json', 'wu.json'] [BOINC] Resolved soft link: ../../projects/axiom.heliex.net/wu_10_g492562_90128_input.json [BOINC] Resolved soft link: ../../projects/axiom.heliex.net/wu_10_g492562_90128_0_r1730012341_0 [BOINC] Actual WU file: ../../projects/axiom.heliex.net/wu_10_g492562_90128_input.json [BOINC] Successfully loaded work unit [BOINC] Expert: 10 [BOINC] Run duration: 150 [BOINC] Sync interval: 999999999 [Trainer] Initializing expert 10... [Trainer] Expert 10 initialized: 42,580,994 parameters [Trainer] Memory usage: ~162.4 MB [GPU] Auto batch size: 128 (1739MB free VRAM, OOM-adaptive) [GPU] No mid-task sync (crunch mode) [GPU] Multi-expert: 3 experts [10, 417, 237] (487MB weights) [DataProducer] Reading from: /var/lib/boinc/Axiom/contribute [Trainer] Starting Hebbian training loop... [Trainer] Learning rate: 0.001 [Trainer] Weight decay: 0.999 [Trainer] Mini-batch size: 128 [Trainer] Sync interval: 999999999 samples [GPU] Pre-loading all data for crunch mode... [DataProducer] Preloaded 101 files -> 4,012,400 samples (501,558 bytes) [GPU] Crunching 4,012,400 samples, batch=128, 3 expert(s) round-robin [GPU] OOM! Batch halved to 64 [GPU] OOM! Batch halved to 32 [WeightSync] Connecting to ws://65.21.196.61:8765... [WeightSync] Connected! [GPU] Expert 10: 2,848 samples [GPU] Expert 417: 2,816 samples [GPU] Expert 237: 2,816 samples [BOINC] Training stopped. 8,480 total samples [BOINC] CUDA synced, memory freed [BOINC] Compressed expert 10 (delta=0.1s, compress=0.6s) [BOINC] Compressed expert 417 (delta=0.1s, compress=0.4s) [BOINC] Compressed expert 237 (delta=0.1s, compress=0.2s) [BOINC] Flushed expert 10: 2826 samples [BOINC] Flushed expert 417: 2826 samples [BOINC] Flushed expert 237: 2826 samples [Sync] Worker-0: Received peer delta (15% merge) [BOINC] Timeout waiting for tasks [BOINC] Complete: syncs=0, samples=8480 2026-02-07 08:47:34 (3076402): axiom_worker_3.41_cuda_x86_64-pc-linux-gnu exited; CPU time 165.299358 2026-02-07 08:47:34 (3076402): called boinc_finish(0) </stderr_txt> ]]>
Network Statistics
38
Contributors
11,887,650
Total Syncs
420
Experts
151,326,327
Credits
© 2026 Axiom Project 2026