me/env-ansible
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

llm, edit provider suffixes.

Browse Source
This commit is contained in:
me
2026-10-15 12:07:56 -07:00
parent 490df1e24c
commit 0f4fe291b2
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tasks/workstation/linux/software/llm.yml
+6 -6
View File
@@ -113,7 +113,7 @@
llm_removals:
##### QWEN #####
- { "purpose": ["chat"], "model": "qwen3:4b", "provider": "qwen_30_0040", "size_mb": 4000, "context": 123456 } # succeeded
- { "purpose": ["chat", "code"], "model": "qwen3.6:35b", "provider": "qwen_36_0350", "size_mb": "{{ llm_size_coef * 23000 }}", "context": 262144 } # succeeded, not dense so not recommended
- { "purpose": ["chat", "code"], "model": "qwen3.6:35b", "provider": "qwen_36_0350_moe", "size_mb": "{{ llm_size_coef * 23000 }}", "context": 262144 } # succeeded, not dense so not recommended
##### Mistral #####
- { "purpose": ["code"], "model": "devstral:24b", "provider": "mistral_dev10_0240", "size_mb": "{{ llm_size_coef * 14000 }}", "context": 128000 } # succeeded
# https://ollama.com/library/mistral-small : Last Updated 2025-01-30, succeeded
@@ -139,8 +139,8 @@
- { "purpose": ["chat"], "model": "qwen3.5:4b", "provider": "qwen_35_0040", "size_mb": "{{ llm_size_coef * 3300 }}", "context": 256000 }
- { "purpose": ["chat"], "model": "qwen3.5:9b", "provider": "qwen_35_0090", "size_mb": "{{ llm_size_coef * 6600 }}", "context": 256000 }
- { "purpose": ["chat", "code"], "model": "qwen3.5:27b", "provider": "qwen_35_0270", "size_mb": "{{ llm_size_coef * 17000 }}", "context": 256000 } # succeeded
- { "purpose": ["chat", "code"], "model": "qwen3.5:35b", "provider": "qwen_35_0350", "size_mb": "{{ llm_size_coef * 22000 }}", "context": 256000 }
- { "purpose": ["chat", "code"], "model": "qwen3.5:122b", "provider": "qwen_35_1220", "size_mb": "{{ llm_size_coef * 81000 }}", "context": 256000 }
- { "purpose": ["chat", "code"], "model": "qwen3.5:35b", "provider": "qwen_35_0350_moe", "size_mb": "{{ llm_size_coef * 22000 }}", "context": 256000 }
- { "purpose": ["chat", "code"], "model": "qwen3.5:122b", "provider": "qwen_35_1220_moe", "size_mb": "{{ llm_size_coef * 81000 }}", "context": 256000 }
### QWEN 3.0 Coder - Supposed to be better than 2.5 for machines that can handle it.
- { "purpose": ["code", "chat"], "model": "qwen3-coder:30b", "provider": "qwen_30_coder_0300", "size_mb": "{{ llm_size_coef * 19000 }}", "context": 123456 }
- { "purpose": ["code", "chat"], "model": "qwen3-coder:420b", "provider": "qwen_30_coder_4200", "size_mb": "{{ llm_size_coef * 290000 }}", "context": 123456 }
@@ -860,14 +860,14 @@
{% if llm.model in llm_models.stdout %}
{% if "code" in llm.purpose %}
- name: {{ llm.provider }}_chat-edit
- name: {{ llm.provider }}_CHAT
provider: ollama
model: {{ llm.model }}
defaultCompletionOptions:
contextLength: {{ [llm_context, llm.context, 32000] | min | int }}
maxTokens: {{ [llm_tokens, (llm.context / 4), 8000] | min | int }}
- name: {{ llm.provider }}_autocomplete
- name: {{ llm.provider }}_AUTO
provider: ollama
model: {{ llm.model }}
defaultCompletionOptions:
@@ -902,7 +902,7 @@
{
"name": "{{ llm.provider
}}_{{ 'CODE' if "code" in llm.purpose else 'CHAT'
}}{{ '' if llm.model in llm_models.stdout else '_MISSING'
}}{{ '' if llm.model in llm_models.stdout else '-MISSING'
}}"
,"baseUrl": "http://127.0.0.1:11434/v1"
,"envKey": "OLLAMA_API_KEY"