Qwopus-GLM-18B with Hermes Agent

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

Task completion

90% (27/30)

Tool-call success

68.9%

Decode tok/s

19.9

Prefill tok/s

1451.3

Server memory

13.1 GB

Avg turns

3.6

Setup

Model: Qwopus-GLM-18B
Family / maker: Community merge · KyleHessling1 (community)
Parameters: 18B
Architecture: dense (merge)
Weights: KyleHessling1/Qwopus-GLM-18B-Merged-GGUF
Quant: Q4_K_M (GGUF, healed)
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: ≤16 GB (measured)

Tested file: Qwopus-GLM-18B-Healed-Q4_K_M.gguf (the healed merge, the only Q4_K_M in the repo). Task t09 (~127K-token prompt with this tokenizer) ran at 192K context.

Per-task results

t01: 3/3
t02: 2/3
t03: 3/3
t04: 3/3
t05: 3/3
t06: 3/3
t07: 3/3
t08: 3/3
t09: 1/3
t10: 3/3

Copy-paste config

1. Start the model server (llama.cpp)

Exact command used in our test. Download the GGUF from KyleHessling1/Qwopus-GLM-18B-Merged-GGUF into ./models/.

llama-server --model ./models/Qwopus-GLM-18B-Healed-Q4_K_M.gguf --alias qwopus-glm-18b --host 127.0.0.1 --port 18080 --ctx-size 196608 --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 qwopus-glm-18b