All issues

Search results

On Windows the engine sizes its pools as if no other process held any of the card

windows

Status: Open.

#570 In FlashML-org/FreeToken;

· YevheniiKotyrlo opened on Sep 29, 2026

A hybrid model's cached prefix cannot be resumed far back: no GDN snapshot survives below a prompt's last chunk

windows

Status: Open.

#569 In FlashML-org/FreeToken;

· YevheniiKotyrlo opened on Sep 29, 2026

Serving Qwen3.8-Flash-Next's 1M-token window on one 24 GB card

Status: Open.

#568 In FlashML-org/FreeToken;

· YevheniiKotyrlo opened on Sep 29, 2026

A stream that is silent through a long prefill is dropped by Node clients and prefilled for nobody

windows

Status: Open.

#567 In FlashML-org/FreeToken;

· YevheniiKotyrlo opened on Sep 29, 2026

QSA block top-k runs a 1M-token window on one program per row

windows

Status: Open.

#566 In FlashML-org/FreeToken;

· YevheniiKotyrlo opened on Sep 29, 2026

Bug: AssertionError: Not enough memory for KV cache when serving Qwen3.8-27B-NVFP4 on Windows

bugSomething isn't workingSomething isn't working

windows

Status: Open.

#561 In FlashML-org/FreeToken;

· hulonghe opened on Sep 28, 2026

Feature: --api-key flag for serve mode (token auth on the HTTP API)

Status: Open.

#557 In FlashML-org/FreeToken;

· santanu20 opened on Sep 27, 2026

[Desktop] weights failed to load: Qwen3.6-27B NVFP4

new-model

Status: Open.

#556 In FlashML-org/FreeToken;

· vincentdapoliza-cmyk opened on Sep 27, 2026

[Desktop] weights failed to load: Gemma-4-26B-A4B NVFP4

new-model

Status: Open.

#555 In FlashML-org/FreeToken;

· lazyper0304 opened on Sep 27, 2026

[Feature] Cumulative prefill/decode seconds on /v1/stats so pollers can compute speeds

Status: Open.

#552 In FlashML-org/FreeToken;

· mgeldi opened on Sep 25, 2026

[Desktop] weights failed to load: occamy-1.0 NVFP4

new-model

Status: Open.

#550 In FlashML-org/FreeToken;

· MaxManMax482 opened on Sep 25, 2026

Qwen3.8-Flash-Next-GSQ-RCO-GGUF not loading.

bugSomething isn't workingSomething isn't working

linux

Status: Open.

#549 In FlashML-org/FreeToken;

· DarkReaperBoy opened on Sep 25, 2026