VERIFIEDNov 222026
Is Raw Phi-4 mini direct logits state of the art?#386Best current rank: #386 in Decision models, as of Nov 22, 2026.
As of Nov 22, 2026, Raw Phi-4 mini direct logits is ranked #386 in Decision models (since Nov 22, 2026; sources JevBench; methodology v1.0). Ranked #386 in Decision models.
At a glance
Facts and scores
Raw Phi-4 mini direct logits is an open-weights (MIT) text model from Microsoft; no public parameter count and no public price.
How it scores
- On JevBench price it scores 3.6%, #81 of 134 (top 61%); 3.5 pts behind verdict-small (Manavarya09, multilingual-e5-small 118M); Jev 1.13.0 (TypeSafe AI) is 0.4 pts above and SPX-CD Flash (SurdAI, hosted /v1/systemone, Oct-4 checkpoint) is 0.3 pts below.
- On JevBench it scores 35.3 (CI 34.8–35.8), #115 of 134 (top 86%); 47.5 behind Perplexity Decider v1.1 27B (Qwen3.8-27B, noncausal decision readout); open-jev-deberta-v3-large (local CPU) is 0.1 above and Raw Qwen3 4B Instruct 2507 direct logits is 0.7 below.
History
First result on isitsota: Oct 8, 2026. Best overall rank so far: #386 in Decision models, first held Nov 22, 2026.
$0.0363 per 1,000 decisions.
Receipts
Evidence
| Source | Score | Date | Type | Receipt |
|---|---|---|---|---|
| JevBench Raw Phi-4 mini direct logits | 35.3 34.8–35.8 | 2026-10-08 | INDEP. | snap_jevbench_2026-10-08_v1.6.1|jevbench-v1.6.1 |
Reign history
Where it sits in the succession
Crown intervals for this model are still empty.
Honest about the gaps
Where it is not the crown
| Category | This model | Rank | Leader |
|---|---|---|---|
| Decision models | #386 in this category | #386 | karpo-decision-31B |
Links
Related pages
Is Raw Phi-4 mini direct logits SOTA?
Best current rank: #386 in Decision models, as of Nov 22, 2026.
What is Raw Phi-4 mini direct logits best at?
As of Nov 22, 2026, Raw Phi-4 mini direct logits is ranked #386 in Decision models (since Nov 22, 2026; sources JevBench).