ghostnet.ai/infrastructure
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

LLM-Server: Testing nach bestandener 64k-Abnahme auf 128k-Zielkontext gestellt

Browse Source 
inference_engine_ctx_size_override auf null (GNAI-008 7.5, GNAI-010 10.3);
131072 aktiv seit 01.11.2026, kein OOM (VRAM 14,3/14,7 GiB von je 16 GiB).

Generated by Mistral Vibe.
Co-Authored-By: Mistral Vibe <vibe@mistral.ai>
This commit is contained in:
spektr
2026-11-01 17:32:17 +01:00
parent 3306b3de4b
commit ca79cc404a
1 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
llm-server/llm/ansible/inventories/testing/group_vars/llm_servers.yml
+5 -3
View File
@@ -9,6 +9,8 @@ inference_engine_user_home: /var/lib/ghostnet-ai/testing
inference_engine_unit_prefix: gnai-testing-llama
inference_engine_network_name: gnai-testing-llama
# GNAI-008 section 7.5: bring-up starts at 64k; the 128k target profile is
# accepted separately (GNAI-010 section 10.3). Set to null after acceptance.
inference_engine_ctx_size_override: 65536
# GNAI-008 section 7.5: bring-up started at 64k; the 128k target profile is
# accepted separately (GNAI-010 section 10.3). The 64k acceptance passed on
# 31.10.2026 (GNAI-010 section 10.5); null activates the profile target
# context of 131072.
inference_engine_ctx_size_override: null