Jared PalmerCloudflareDecision matchup
Kev 4Bvs.Clef
Which model makes better decisions?
Clef is slightly ahead in independent tests, but the lead is not certain.
Pick Kev 4B for
Pick Clef for
Shared tests
Independent evaluators and leaderboards only, the newest shared snapshot per evaluator. With 95 % ranges a lead counts when they do not overlap; for hit rates, when it holds in a 95 % test. Without any uncertainty figure at most “slightly ahead”.
AIMultiple decision-models evaluation
Clef slightly aheadClef is clearly ahead in 0 of 4 subtests and slightly in 3, Kev 4B in 0 clearly and slightly in 1.
AIMultiple decision-task benchmark: Task done check
as of October 18, 2026
AIMultiple decision-task benchmark: Research paper topic
500 tasks, as of October 18, 2026
AIMultiple decision-task benchmark: Security flaw severity
500 tasks, as of October 18, 2026
Show all 4 subtests
| Subtest | Kev 4B | Clef | Verdict |
|---|---|---|---|
| AIMultiple decision-task benchmark: Task done check | 87.40 | 99.50 | Clef slightly ahead |
| AIMultiple decision-task benchmark: Research paper topic | 66.40 | 70.20 | Clef slightly ahead |
| AIMultiple decision-task benchmark: Security flaw severity | 40.40 | 42.70 | Clef slightly ahead |
| AIMultiple decision-task benchmark: Recall danger level | 61.10 | 59.20 | Kev 4B slightly ahead |
Source · latest October 18, 2026
All data side by side
| Kev 4B | Clef | |
|---|---|---|
| Kind | Adapter | Standalone |
| Developer | Jared Palmer | Cloudflare |
| Status | Open source | Active |
| Released | September 20, 2026 | October 1, 2026 |
| Weights | open | open |
| Runs locally | yes | yes |
| Parameters | n/a | 27 B |
| License | Apache-2.0 for the adapter and head | Apache-2.0 |
| Price | n/a | $0.240 per 1M input tokens (21,818 neurons per 1M) |
| Popularity (out of 100) | 23 | 51 |