Skip to content

Actions: FlashML-org/FreeToken

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
232 workflow runs
232 workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Nightly wheels Nightly wheels #61: Scheduled
3m 33s main
main
3m 33s
GLM-5.3-Flash-NVFP4 CUDA OOM during load on RTX 5070 Ti 16 GB (sm_120) Label issues #154: Issue #611 opened by ogino
6s
6s
Windows wheels resolve unpinned torch dependency to CPU-only build: 'cannot use CUDA device 0: only 0 device(s) visible' Label issues #153: Issue #607 opened by pablosilva87
7s
7s
DeepSeek-V4-Flash-0731-NVFP4 (nvidia) fails on 0.1.3+gfa0e9537d: WeightLoadError KeyError 'layers.0.ffn.experts.0.w1.scale' Label issues #152: Issue #606 opened by pablosilva87
8s
8s
Nightly wheels Nightly wheels #60: Scheduled
3m 50s main
main
3m 50s
Promote nightly to beta Promote nightly to beta #3: Manually run by jason-fxz
29s main
main
29s
FreeToken Roadmap (2026) Label issues #151: Issue #79 edited by jason-fxz
7s
7s
Nightly wheels Nightly wheels #59: Scheduled
6s main
main
6s
[Engine] /v1/stats reports frontend default instead of resolved backend KV page size Label issues #150: Issue #598 edited by yunwei37
8s
8s
[Engine] Qwen3.8-Flash-Next-FP8 exhausts host RAM on RTX 5090 + 128 GB after serial loading fallback Label issues #149: Issue #600 opened by yunwei37
10s
10s
[Engine] /v1/stats reports frontend default instead of resolved backend KV page size Label issues #148: Issue #598 opened by yunwei37
6s
6s
[Windows][Desktop] WeightLoadError: invalid python storage with Qwen3.6 NVFP4 and gpt-oss-20b MXFP4 Label issues #147: Issue #597 opened by Shahrad-gh
8s
8s
Nightly wheels Nightly wheels #58: Scheduled
6s main
main
6s
WeChat Group QRCode Label issues #146: Issue #482 edited by jason-fxz
6s
6s
Nightly wheels Nightly wheels #57: Scheduled
3m 31s main
main
3m 31s
Nightly wheels Nightly wheels #56: Scheduled
3m 48s main
main
3m 48s
The computer on the local area network cannot be accessed Label issues #145: Issue #584 opened by albertjacksoon
16s
16s
An offloaded prefill streams every expert of each layer, however short the chunk Label issues #144: Issue #579 opened by YevheniiKotyrlo
9s
9s
Nightly wheels Nightly wheels #55: Scheduled
3m 29s main
main
3m 29s
Serving Qwen3.8-Flash-Next's 1M-token window on one 24 GB card Label issues #143: Issue #568 edited by YevheniiKotyrlo
10s
10s
QSA block top-k runs a 1M-token window on one program per row Label issues #142: Issue #566 edited by YevheniiKotyrlo
7s
7s
On Windows the engine sizes its pools as if no other process held any of the card Label issues #141: Issue #570 opened by YevheniiKotyrlo
9s
9s
A hybrid model's cached prefix cannot be resumed far back: no GDN snapshot survives below a prompt's last chunk Label issues #140: Issue #569 opened by YevheniiKotyrlo
8s
8s
Serving Qwen3.8-Flash-Next's 1M-token window on one 24 GB card Label issues #139: Issue #568 opened by YevheniiKotyrlo
10s
10s
A stream that is silent through a long prefill is dropped by Node clients and prefilled for nobody Label issues #138: Issue #567 opened by YevheniiKotyrlo
10s
10s