mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-05-03 02:11:48 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

swe-smith-oracle runs 1 step process. llama server was just breaking again locally idk, works through Hermes endpoint & ManagedServer fine

Browse source
This commit is contained in:
Shannon Sands 2026-02-04 11:22:45 +10:00
parent 6cb4fe948a
commit 5a9c98a771
8 changed files with 359 additions and 51 deletions
Download patch file Download diff file Expand all files Collapse all files

6
.env.example
View file

@ -28,6 +28,12 @@ HERMES_BACKEND=openai
# ATROPOS_TOKENIZER_NAME=NousResearch/Hermes-4.3-36B
# ATROPOS_SERVER_API_KEY=local
#
# Hosted Nous inference API:
# ATROPOS_SERVER_BASE_URL=https://inference-api.nousresearch.com
# ATROPOS_SERVER_MODEL=Hermes-4.3-36B
# ATROPOS_TOKENIZER_NAME=NousResearch/Hermes-4.3-36B
# ATROPOS_SERVER_API_KEY=sk-... (Bearer token)
#
# If you plan to run GRPO-style group sampling (e.g. `--env.group_size 4`) against
# llama.cpp, start the server with at least that many slots, e.g.:
# LLAMA_CPP_PARALLEL=4 Hermes-Agent/scripts/launch_llama_cpp_hermes_4_36b.sh