Adriano/Multi_Swarm_Coevolutive
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

revert(config): rollback tier C a qwen-2.5-72b-instruct (qwen3-235b inferiore)

Browse Source 
Run controllo phase2-qwen25-control-001 (seed 42, stessa pipeline Phase 2,
solo tier C switched) ha dimostrato che qwen-2.5-72b è qualitativamente
SUPERIORE a qwen3-235b sul nostro workload:

| metrica           | qwen3-235b | qwen-2.5-72b | delta |
| ----------------- | ---------- | ------------ | ----- |
| max fitness       | 0.0238     | 0.0311       | +30%  |
| median > 0 in gen | mai        | 4 gen su 10  | --    |
| entropy media     | 0.199      | 0.85         | 4.3x  |
| genomi fit > 0    | 5          | 10           | 2x    |
| parse success     | 97.7%      | 100%         | +     |
| durata            | 50 min     | 28 min       | 0.56x |
| LLM calls         | 148        | 90           | 0.61x |
| cost USD          | 0.0223     | 0.0122       | 0.55x |

Controintuitivo: 235B con context 262k era atteso superiore al 72B legacy.
In pratica qwen3-235b in tier C produce strategie meno diverse,
meno parsabili e meno ottimizzabili dal GA.

Ripristinati prezzi cost_tracker tier C a 0.40/0.40 (qwen-2.5-72b).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
AdrianoDev
2026-05-11 23:45:52 +02:00
parent 9344395760
commit 8ec45c5c1b
7 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/multi_swarm/llm/client.py
+1 -1
View File
@@ -17,7 +17,7 @@ from ..genome.hypothesis import HypothesisAgentGenome, ModelTier
MODEL_TIER_S = "google/gemini-3-flash-preview"
MODEL_TIER_A = "deepseek/deepseek-v4-flash"
MODEL_TIER_B = "deepseek/deepseek-v4-flash"
MODEL_TIER_C = "qwen/qwen3-235b-a22b-2507"
MODEL_TIER_C = "qwen/qwen-2.5-72b-instruct"
MODEL_TIER_D = "openai/gpt-oss-20b"
OPENROUTER_BASE_URL = "https://openrouter.ai/api/v1"