Menu

← Leaderboard

Hermes-4-70B with Hermes Agent

Q4_K_M (GGUF) on llama.cpp commit a37cc46. Tested

Task completion
70% (21/30)
Tool-call success
65.5%
Decode tok/s
4.5
Prefill tok/s
361.3
Server memory
59.5 GB
Avg turns
4.6

Setup

Model
Hermes-4-70B
Family / maker
Hermes · Nous Research
Parameters
70B
Architecture
dense
Weights
unsloth/Hermes-4-70B-GGUF
Quant
Q4_K_M (GGUF)
Engine
llama.cpp commit a37cc46
Hermes Agent
v0.22.4
Tested
2026-10-23
Context
64K (65,536 tokens); long-context task t09 at 128K or 192K
Memory tier
≤128 GB (measured)

Per-task results

  • t01: 3/3
  • t02: 2/3
  • t03: 3/3
  • t04: 2/3
  • t05: 3/3
  • t06: 2/3
  • t07: 3/3
  • t08: 0/3
  • t09: 0/3
  • t10: 3/3

Copy-paste config

1. Start the model server (llama.cpp)

Exact command used in our test. Download the GGUF from unsloth/Hermes-4-70B-GGUF into ./models/.

llama-server --model ./models/Hermes-4-70B-Q4_K_M.gguf --alias hermes-4-70b --host 127.0.0.1 --port 18080 --ctx-size 131072 --parallel 1 --n-gpu-layers 999 --flash-attn on --jinja

2. Point Hermes Agent at it

hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:18080/v1
hermes config set model.default hermes-4-70b

Other Hermes setups