diff --git a/recipes/torchcam/meta.yaml b/recipes/torchcam/meta.yaml
new file mode 100644
index 0000000000000..899c629cda3f9
--- /dev/null
+++ b/recipes/torchcam/meta.yaml
@@ -0,0 +1,55 @@
+{% set name = "torchcam" %}
+{% set version = "0.3.1" %}
+
+
+package:
+  name: {{ name|lower }}
+  version: {{ version }}
+
+source:
+  url: https://pypi.io/packages/source/{{ name[0] }}/{{ name }}/torchcam-{{ version }}.tar.gz
+  sha256: 400d451cbbafe0dbddbf3a22feab9e23d1317e90cf6eae16bca96e0e2ff4f076
+
+build:
+  number: 0
+  noarch: python
+  script:     
+    ## Ensure exclusion of tests folder
+    - rm -rf test  # [not win]
+    ## Install package {{ name }} from PyPI
+    - {{ PYTHON }} -m pip install . -vv
+
+requirements:
+  host:
+    - pip
+    - python >=3.6
+  run:
+    - python >=3.6
+    - matplotlib-base >=3.0.0
+    - numpy >=1.14.0
+    - pillow >=8.3.2    
+    - pytorch >=1.5.1
+
+test:
+  imports:
+    - torchcam
+    - torchcam.methods
+  commands:
+    - pip check
+  requires:
+    - pip
+
+about:
+  home: https://github.com/frgfm/torch-cam
+  summary: Class activation maps for your PyTorch CNN models
+  license: Apache-2.0
+  license_file: LICENSE
+  description: |
+    Simple way to leverage the class-specific activation of 
+    convolutional layers in PyTorch.
+  doc_url: https://frgfm.github.io/torch-cam
+  dev_url: https://github.com/frgfm/torch-cam
+
+extra:
+  recipe-maintainers:
+    - sugatoray