Mirrors/ProxMenux
1
0
Fork 0
mirror of https://github.com/MacRimi/ProxMenux.git synced 2026-08-03 06:16:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

update version 1.2.3

Browse Source
This commit is contained in:
MacRimi
2026-07-13 23:19:24 +02:00
parent 1b623cd275
commit 1872a309ec
7 changed files with 60 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files
AppImage/config/verified_ai_models.json
+26 -23
View File
@@ -1,30 +1,30 @@
{
"_description": "Verified AI models for ProxMenux notifications. Only models listed here will be shown to users. Models are tested to work with the chat/completions API format.",
"_updated": "2026-04-19",
"_updated": "2026-07-13",
"_verifier": "Refreshed with tools/ai-models-verifier (private). Re-run before each ProxMenux release to keep the list current. The verifier and ProxMenux share the same reasoning/thinking-model handlers so their verdicts stay aligned with runtime behaviour.",
"groq": {
"models": [
"llama-3.3-70b-versatile",
"llama-3.1-70b-versatile",
"llama-3.1-8b-instant",
"llama3-70b-8192",
"llama3-8b-8192",
"mixtral-8x7b-32768",
"gemma2-9b-it"
"openai/gpt-oss-120b",
"openai/gpt-oss-20b"
],
"recommended": "llama-3.3-70b-versatile",
"_note": "Not yet re-verified in 2026-04 refresh — kept from previous curation. Run the verifier with a Groq key to prune deprecated entries."
"_note": "Verified 2026-07-13 against Groq's public production catalog. Legacy llama-3.1-70b-versatile / llama3-70b-8192 / llama3-8b-8192 / mixtral-8x7b-32768 / gemma2-9b-it removed (retired upstream). openai/gpt-oss-120b (500 T/s) and openai/gpt-oss-20b (1000 T/s) added — both listed as current production."
},
"gemini": {
"models": [
"gemini-flash-lite-latest",
"gemini-2.5-flash-lite",
"gemini-2.5-flash",
"gemini-3-flash-preview"
"gemini-3.1-flash-lite",
"gemini-3-flash-preview",
"gemini-3.5-flash"
],
"recommended": "gemini-2.5-flash-lite",
"_note": "flash-lite / flash pass the verifier consistently; pro variants reject thinkingBudget=0 and are overkill for notification translation anyway. 'latest' aliases (gemini-flash-latest, gemini-flash-lite-latest) are intentionally omitted because they resolved to different models across runs and produced timeouts in some regions.",
"_note": "Verified 2026-07-13. gemini-flash-lite-latest now passes consistently (1.6s) and is fastest, but gemini-2.5-flash-lite remains recommended because 'latest' aliases can drift over time. gemini-3.1-flash-lite is the stable successor to 3-flash-preview. Pro variants continue to reject thinkingBudget=0 and are overkill for notification translation.",
"_deprecated": ["gemini-2.0-flash", "gemini-2.0-flash-lite", "gemini-1.5-flash", "gemini-1.0-pro", "gemini-pro"]
},
@@ -36,21 +36,23 @@
"gpt-4.1",
"gpt-4o",
"gpt-5-chat-latest",
"gpt-5.4-nano",
"gpt-5.4-mini"
"gpt-5-nano"
],
"recommended": "gpt-4.1-nano",
"_note": "Reasoning models (o-series, gpt-5/5.1/5.2 non-chat variants) are supported by openai_provider.py via max_completion_tokens + reasoning_effort=minimal, but not listed here by default: their latency is higher than the chat models and they do not improve translation quality for notifications. Add specific reasoning IDs to this list only if a user explicitly wants them."
"_note": "Verified 2026-07-13. gpt-5.4-nano / gpt-5.4-mini removed (HTTP 400 — provider params rejected). gpt-5-nano added (2.0s, current-gen fast). Reasoning models (o-series, gpt-5/5.1/5.2 non-chat variants) are supported by openai_provider.py via max_completion_tokens + reasoning_effort=minimal, but not listed here: their latency is higher and they do not improve translation quality for notifications. Add specific reasoning IDs to this list only if a user explicitly wants them."
},
"anthropic": {
"models": [
"claude-3-5-haiku-latest",
"claude-3-5-sonnet-latest",
"claude-3-opus-latest"
"claude-haiku-4-5",
"claude-sonnet-5",
"claude-opus-4-8",
"claude-sonnet-4-6",
"claude-opus-4-6",
"claude-fable-5"
],
"recommended": "claude-3-5-haiku-latest",
"_note": "Not re-verified in 2026-04 refresh — kept from previous curation. Add claude-4.x / claude-4.5 / claude-4.6 / claude-4.7 variants after running the verifier with an Anthropic key."
"recommended": "claude-haiku-4-5",
"_note": "Verified 2026-07-13 with all 10 discovered models passing after aligning the verifier with anthropic_provider.py (temperature omitted — newest generations reject it with 'temperature is deprecated for this model'). Legacy claude-3-5-haiku-latest / claude-3-5-sonnet-latest / claude-3-opus-latest removed (deprecated upstream, not in the Models API). haiku-4-5 is the sweet spot for notification translation (3.6s, $1/$5 per MTok); sonnet-5 for slightly richer output (3.1s, $3/$15); opus-4-8 / fable-5 for demanding cases."
},
"openrouter": {
@@ -58,15 +60,16 @@
"meta-llama/llama-3.3-70b-instruct",
"meta-llama/llama-3.1-70b-instruct",
"meta-llama/llama-3.1-8b-instruct",
"anthropic/claude-3.5-haiku",
"anthropic/claude-3.5-sonnet",
"google/gemini-flash-1.5",
"meta-llama/llama-4-scout",
"anthropic/claude-haiku-4.5",
"anthropic/claude-sonnet-4.6",
"google/gemini-2.5-flash-lite",
"google/gemini-2.5-flash",
"openai/gpt-4o-mini",
"mistralai/mistral-7b-instruct",
"mistralai/mixtral-8x7b-instruct"
"mistralai/mistral-small-3.2-24b-instruct"
],
"recommended": "meta-llama/llama-3.3-70b-instruct",
"_note": "Not re-verified in 2026-04 refresh. google/gemini-flash-2.5-flash-lite was malformed in the previous entry and has been replaced with google/gemini-flash-1.5."
"_note": "Verified 2026-07-13 against OpenRouter's public /api/v1/models catalog (343 models advertised). anthropic/claude-3.5-haiku / anthropic/claude-3.5-sonnet / google/gemini-flash-1.5 / mistralai/mistral-7b-instruct / mistralai/mixtral-8x7b-instruct removed (GONE from catalog). Modern replacements added: llama-4-scout (Meta's current gen), claude-haiku-4.5 / claude-sonnet-4.6 (Anthropic current gen via OpenRouter proxy), gemini-2.5-flash / flash-lite (Google current gen), mistral-small-3.2-24b (Mistral current small)."
},
"ollama": {