Search results

<code>Llama.embed</code> crashes when <code>n_batch</code> &gt; 512

bugSomething isn't workingSomething isn't working

Status: Open.

#1762 In abetlen/llama-cpp-python;

· lsorber opened on Sep 25, 2024

No matter how many times I build it, it won't start

bugSomething isn't workingSomething isn't working

build

Status: Open.

#1719 In abetlen/llama-cpp-python;

· Enchante503 opened on Aug 31, 2024

flash attention on Nvidia Tesla P100s results in the <code>CUDA error: unspecified launch failure</code> - (<code>CUDA kernel flash_attn_tile_ext_f16 has no device code compatible with CUDA arch 520</code>)

bugSomething isn't workingSomething isn't working

build

Status: Open.

#1710 In abetlen/llama-cpp-python;

· AlHering opened on Aug 26, 2024

Problem with v0.2.88

bugSomething isn't workingSomething isn't working

Status: Open.

#1684 In abetlen/llama-cpp-python;

· atrisovic opened on Aug 14, 2024

High CPU Usage, very slow performance, with flash_attn=true on ROCM 6.1.2

bugSomething isn't workingSomething isn't working

Status: Open.

#1661 In abetlen/llama-cpp-python;

· curvedinf opened on Aug 6, 2024

Error When Loading Model with llama_cpp: [WinError -1073741795] Windows Error 0xc000001d

bugSomething isn't workingSomething isn't working

Status: Open.

#1660 In abetlen/llama-cpp-python;

· ltkien2003 opened on Aug 6, 2024

pip install llama-cpp-python on anaconda

bugSomething isn't workingSomething isn't working

Status: Open.

#1654 In abetlen/llama-cpp-python;

· werruww opened on Aug 5, 2024

Installing the pre-built v0.2.85 – v0.3.0 Metal wheels for Python 3.9 – 3.12 yields <code>Bad magic number for file header</code>

bugSomething isn't workingSomething isn't working

Status: Open.

#1650 In abetlen/llama-cpp-python;

· lsorber opened on Aug 2, 2024

create_chat_completion is stuck in versions 0.2.84 and 0.2.85 for Mac Silicon

bugSomething isn't workingSomething isn't working

Status: Open.

#1648 In abetlen/llama-cpp-python;

· mobeetle opened on Aug 1, 2024

All requests end with 'finish_reason': 'length' when the max_tokens=-1 parameter is set.

bugSomething isn't workingSomething isn't working

Status: Open.

#1645 In abetlen/llama-cpp-python;

· tur0kmagalp opened on Jul 31, 2024

Pre-built cpu wheel does not work on Ubuntu due to libc.musl dependency

bugSomething isn't workingSomething isn't working

Status: Open.

#1628 In abetlen/llama-cpp-python;

· OKUA1 opened on Jul 27, 2024

CUDA error: unspecified launch failure on inference on Nvidia V100 GPUs

bugSomething isn't workingSomething isn't working

Status: Open.

#1624 In abetlen/llama-cpp-python;

· rplescia opened on Jul 26, 2024