Search results

Can not capture graph with torch.compile (dynamo)

Status: Open.

#64 In turboderp-org/exllamav3;

· llcnt opened on Jul 21, 2025

Jamba Mini Support

Status: Open.

#63 In turboderp-org/exllamav3;

· Downtown-Case opened on Jul 18, 2025

VRAM usage vs. EXL2

Status: Open.

#60 In turboderp-org/exllamav3;

· cmoncure opened on Jul 9, 2025

Hunyuan-A13B-Instruct support

Status: Open.

#59 In turboderp-org/exllamav3;

· Originalimoc opened on Jul 8, 2025

Another improved quantization method from QTIP team?

Status: Open.

#56 In turboderp-org/exllamav3;

· Originalimoc opened on Jun 27, 2025

GPU split isn't followed here?

Status: Open.

#54 In turboderp-org/exllamav3;

· Ph0rk0z opened on Jun 19, 2025

Command A support

Status: Open.

#53 In turboderp-org/exllamav3;

· MRGRD56 opened on Jun 18, 2025

Error converting moonshotai/Kimi-Dev-72B

Status: Open.

#52 In turboderp-org/exllamav3;

· artus-dev opened on Jun 17, 2025

dots.llm1 support

Status: Open.

#51 In turboderp-org/exllamav3;

· louiehelm opened on Jun 11, 2025

RuntimeError: Insufficient VRAM in split for model and cache despite remaining 1GB VRAM (exllamav3 + tabbyAPI)

Status: Open.

#50 In turboderp-org/exllamav3;

· meigami0 opened on Jun 9, 2025

running the QTIP inference code

Status: Open.

#49 In turboderp-org/exllamav3;

· tsengalb99 opened on Jun 5, 2025

<code>RuntimeError</code> when requesting logprobs with Gemma 3 models

Status: Open.

#48 In turboderp-org/exllamav3;

· samueldemoura opened on Jun 4, 2025