WEBDIENSTE/ai-suite
1
0
Fork 0
mirror of https://github.com/trevorsandy/ai-suite.git synced 2026-09-30 21:45:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

Change Ollama and LLaMA.cpp context size in .env to 4096 from 8192

Browse Source
This commit is contained in:
Trevor SANDY
2026-01-21 10:46:09 +01:00
parent d5fc04dad9
commit f9b4e02047
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.env.example
+2 -2
View File
@@ -213,7 +213,7 @@ OLLAMA_PORT=11434
OLLAMA_HOST=host.docker.internal:11434
# Tuning
OLLAMA_CONTEXT_LENGTH=8192
OLLAMA_CONTEXT_LENGTH=4096
OLLAMA_FLASH_ATTENTION=1
OLLAMA_KV_CACHE_TYPE=q4_0
OLLAMA_MAX_LOADED_MODELS=2
@@ -271,7 +271,7 @@ LLAMACPP_MODEL_PATH=${LLAMACPP_MODELS_DIR}/${LLAMACPP_MODEL_NAME}
LLAMA_ARG_HF_REPO=${LLAMACPP_MODEL_GEMMA_ID}
# Tuning
LLAMA_ARG_CTX_SIZE=8192
LLAMA_ARG_CTX_SIZE=4096
LLAMA_ARG_FLASH_ATTN=1
LLAMA_ARG_N_GPU_LAYERS=0
LLAMA_ARG_THREADS=4