This is a repository based off of https://github.com/DaveBben/esp32-llm, the goal of which is to run a performant and capable language model off of the esp-32.

A lot of the inspiration and changes were taken from that repo, and the main code to run it was taken from this repo: https://github.com/karpathy/llama2.c

To run: idf.py set-target esp32s3 //maybe not necessary? ^

idf.py build
// or, if it's your first time: This is to set up defaults like the flash size, the flash mode, etc. You may need to change sdkconfig.defaults if your esp32 doesn't have 8 mb of flash, for instance. 
idf.py fullclean build -DIDF_SDKCONFIG_DEFAULTS="sdkconfig.defaults" 

python -m esptool --chip esp32s3 -b 460800 --before default_reset --after hard_reset write_flash --flash_mode dio --flash_size detect --flash_freq 80m 0x0 build\bootloader\bootloader.bin 0x8000 build\partition_table\partition-table.bin 0x10000 build\junipyr.bin 0x110000 build\data.bin

To monitor: python %IDF_PATH%\tools\idf_monitor.py -b 460800

To build, flash, and monitor: idf.py build & python -m esptool --chip esp32s3 -b 460800 --before default_reset --after hard_reset write_flash --flash_mode dio --flash_size detect --flash_freq 80m 0x0 build\bootloader\bootloader.bin 0x8000 build\partition_table\partition-table.bin 0x10000 build\junipyr.bin 0x110000 build\data.bin & python %IDF_PATH%\tools\idf_monitor.py -b 460800

Note: both of these will detect the port automatically, though you probably need just one esp32 plugged in at a time.

This is using this 8MB version of the ESP32 from XIAO through seeed studio: https://www.amazon.com/dp/B0DJ6NQFKX