avalan-ai / avalan / 16993270559
100%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 125
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 99.975%. Remained the same
16993270559

push

github

web-flow 
Expose generation cache options (#580)

* Expose generation cache options

* Updating CLI.md with # avalan

```
usage: avalan [-h] [--cache-dir CACHE_DIR] [--subfolder SUBFOLDER]
              [--tokenizer-subfolder TOKENIZER_SUBFOLDER] [--device DEVICE]
              [--parallel {auto,colwise,rowwise,colwise_rep,rowwise_rep,local_colwise,local_rowwise,local,gather,local_packed_rowwise,sequence_parallel,replicate}]
              [--parallel-count PARALLEL_COUNT]
              [--disable-loading-progress-bar] [--hf-token HF_TOKEN]
              [--locale LOCALE]
              [--loader-class {auto,gemma3,gpt-oss,mistral3}]
              [--backend {transformers,mlx,vllm}] [--locales LOCALES]
              [--low-cpu-mem-usage] [--login] [--no-repl] [--quiet] [--record]
              [--revision REVISION] [--skip-hub-access-check] [--verbose]
              [--version]
              [--weight-type {auto,bool,bf16,f16,f32,f64,fp16,fp32,i8,i16,i32,i64,ui8}]
              [--help-full]
              {agent,cache,deploy,flow,memory,model,tokenizer,train} ...

Avalan CLI

positional arguments:
  {agent,cache,deploy,flow,memory,model,tokenizer,train}

options:
  -h, --help            show this help message and exit
  --cache-dir CACHE_DIR
                        Path to huggingface cache hub (defaults to
                        /Users/mariano/.cache/huggingface/hub, can also be
                        specified with $HF_HUB_CACHE)
  --subfolder SUBFOLDER
                        Subfolder inside model repository to load the model
                        from
  --tokenizer-subfolder TOKENIZER_SUBFOLDER
                        Subfolder inside model repository to load the
                        tokenizer from
  --device DEVICE       Device to use (cpu, cuda, mps). Defaults to mps
  --parallel {auto,colwise,rowwise,colwise_rep,rowwise_rep,local_colwise,local_rowwise,local,gather,local_packed_rowwise,sequence_parallel,replicate}
                        Tensor para... (continued)

8097 of 8099 relevant lines covered (99.98%)

2.0 hits per line

Jobs
ID Job ID Ran Files Coverage
1 unit test - 16993270559.1 125
99.98
 GitHub Action Run
Source Files on build 16993270559