All issues

Search results

API access security verification

Status: Open.

#152 In FlashML-org/FreeToken;

· jackrao168 opened on Aug 24, 2026

ft bench bw picks hybrid on Ampere, but offload is 8.3x faster (DeepSeek-V4-Flash, 2x RTX 3090)

Status: Open.

#151 In FlashML-org/FreeToken;

· GlidLov opened on Aug 24, 2026

AssertionError: cache budget too small

Status: Open.

#150 In FlashML-org/FreeToken;

· Lolmc0587 opened on Aug 24, 2026

Ram size detection issue for loading models, and gpu auto detect issue

Status: Open.

#148 In FlashML-org/FreeToken;

· vingaut opened on Aug 24, 2026

An error occurred while loading the local model

Status: Open.

#147 In FlashML-org/FreeToken;

· Sirius2016 opened on Aug 24, 2026

bench_offload_cache_copy.py default sweep aborts on minimax-m2.5-marlin and skips the last four profiles

Status: Open.

#146 In FlashML-org/FreeToken;

· rakhimovv opened on Aug 24, 2026

[Bug]: Desktop UI window close leaves orphaned backend engine process locking port 8000 and GPU VRAM

Status: Open.

#143 In FlashML-org/FreeToken;

· NitishKashyapR opened on Aug 24, 2026

能否支持更多元量化方式

Status: Open.

#142 In FlashML-org/FreeToken;

· Eliauk-Lik opened on Aug 24, 2026

Add support for all major 2026 KV cache compression techniques (TurboQuant, RotorQuant, KIVI, KVQuant, eviction methods)

Status: Open.

#141 In FlashML-org/FreeToken;

· ArNikiTenS-USA opened on Aug 24, 2026

Please add cuda12.6 and cuda12.8 support during installation

Status: Open.

#140 In FlashML-org/FreeToken;

· xgwang78-dev opened on Aug 24, 2026

Qwen3.8-27B-NVFP4 fails to load — RuntimeError: CUDA driver error: device not ready

Status: Open.

#139 In FlashML-org/FreeToken;

· miguel-oltra opened on Aug 24, 2026

请提供Windows 的 WHL,谢谢

Status: Open.

#130 In FlashML-org/FreeToken;

· wodsoe opened on Aug 24, 2026