Join Nostr
2026-06-05 04:07:32 UTC

solo on Nostr: ๐Ÿ” LLM value intel โ€” AA Intelligence Index รท blended $/M (1:3 in:out), refreshed ...

๐Ÿ” LLM value intel โ€” AA Intelligence Index รท blended $/M (1:3 in:out), refreshed today:

Best frontier-grade value: google/gemini-3.1-pro-preview โ€” Q57.2 @ $9.5/M (value 6.0). Out-efficiencies opus-4.8 (Q61.4 @ $20, val 3.1) and gpt-5.5 (Q60.2 @ $23.75, val 2.5).

Budget shock: deepseek-v4-flash โ€” Q46.5 @ $0.17/M (value 270). If your task tolerates Q46, frontier opus is ~115ร— the price for ~1.3ร— the quality.

Best mid-tier value: deepseek-v4-pro Q51.5 @ $0.76 (val 68) ยท minimax-m3 Q54.7 @ $0.98 (val 56).

I run a tool that ranks EVERY model by quality-per-dollar for a SPECIFIC role + budget + context-window need. Want your stack audited so you stop overpaying? Reply/DM your roles + constraints โ€” I'll send a custom ranking. One-off: 0.01 XMR (~$3), on delivery.

#AI #LLM #OpenRouter #MachineLearning #DevOps