All issues

Search results

[Windows] Qwen3.6-35B-A3B-NVFP4 expert loading crashes — torch_cpu.dll access violation (0xc0000005)

Status: Open.

#304 In FlashML-org/FreeToken;

· huangxianzhan opened on Aug 31, 2026

fix(engine): account for attention workspace in cache-pool sizing

Status: Open.

#303 In FlashML-org/FreeToken;

· ralfkuh-lab opened on Aug 30, 2026

feat(server): add a server-side thinking override

Status: Open.

#302 In FlashML-org/FreeToken;

· ralfkuh-lab opened on Aug 30, 2026

fix(server): allow overriding the distributed port

Status: Open.

#301 In FlashML-org/FreeToken;

· ralfkuh-lab opened on Aug 30, 2026

FreeToken-Intel: 35B MoE, live tokens off an Arc Pro B70 (SYCL / XPU port)

Status: Open.

#299 In FlashML-org/FreeToken;

· aangelinsf opened on Aug 30, 2026

Not a bug - crediting the hit/miss split design

Status: Open.

#297 In FlashML-org/FreeToken;

· sangharshadhyeta opened on Aug 30, 2026

[WSL2] H2D copies from pinned memory silently no-op after heavy pinned-alloc churn (driver 591.86, CUDA 13.0 runtime)

Status: Open.

#291 In FlashML-org/FreeToken;

· x00r opened on Aug 30, 2026

[Win 11] freetoken engine install failed

Status: Open.

#290 In FlashML-org/FreeToken;

· UnknowViewer opened on Aug 30, 2026

docs: docs/models.md has no memory column — the expert pool ranges 9.6 GB to 363.8 GB across the listed checkpoints

Status: Open.

#289 In FlashML-org/FreeToken;

· agentatwork opened on Aug 30, 2026

[Bug] Windows + RTX 50 (sm_120): RuntimeError: Rowwise scaling is not currently supported on your device during CUDA graph capture — dense FP8 W8A8 path

Status: Open.

#288 In FlashML-org/FreeToken;

· lordquest opened on Aug 30, 2026

Windows installer marked as malicious

Status: Open.

#284 In FlashML-org/FreeToken;

· SelfMan opened on Aug 30, 2026

Model loading fails with OSError: [Errno 22] Invalid argument when safetensors are stored on eCryptFS

Status: Open.

#282 In FlashML-org/FreeToken;

· afaulkinberry opened on Aug 30, 2026