Search results

Multi-GPU for nvidia

Status: Open.

#123 In antirez/ds4;

· vacekj opened on May 13, 2026

API access security verification

Status: Open.

#120 In antirez/ds4;

· jackrao168 opened on May 13, 2026

q4-imatrix weights fail to load: "expected IQ2_XXS expert tensors"

bugSomething isn't workingSomething isn't working

metal

weights loading or handlingNew feature or requestNew feature or request

Status: Open.

#114 In antirez/ds4;

· aisaacsmitchell opened on May 13, 2026

Question: why does readme recommend using openai-completions api for coding agent usage.

documentationImprovements or additions to documentationImprovements or additions to documentation

Status: Open.

#113 In antirez/ds4;

· njbrake opened on May 13, 2026

[feature] Offload KV cache from GPU (VRAM) to RAM

Status: Open.

#108 In antirez/ds4;

· d3y4n opened on May 12, 2026

ds4 parser vulnerability: user prompt can insert special tokens that are interpreted by the model

Status: Open.

#95 In antirez/ds4;

· pintorem opened on May 12, 2026

Please provide a ModelScope mirror for Chinese users who cannot easily download large models from Hugging Face

Status: Open.

#85 In antirez/ds4;

· cangming009 opened on May 12, 2026

Feature request: Natural Language Level Speculative Decoding with heterogeneous Draft Models (e.g. Qwen 2B)

help wantedExtra attention is neededExtra attention is needed

speed

wontfixThis will not be worked onThis will not be worked on

Status: Open.

#80 In antirez/ds4;

· vagrillo opened on May 11, 2026

FP4 inference for 4 bit

correctness

Status: Open.

#74 In antirez/ds4;

· antirez opened on May 11, 2026

can you add deepseek q8 ?

Status: Open.

#64 In antirez/ds4;

· jonipupet opened on May 10, 2026

Speculative Decoding via "Draft KV Stitching"

Status: Open.

#63 In antirez/ds4;

· Nitya-003 opened on May 10, 2026

Problems when launching from other programs like llama-swap

Status: Open.

#49 In antirez/ds4;

· links486 opened on May 10, 2026