← Models
phi-4-reasoning-plus
evaluatedVendor: MicrosoftQuant: Q8Type: local
Capability starfish
green = capability (pass-rate) · blue = operational (vs roster) · dashed = roster average
Overall
mean domain rate
Solved
19/48
latest per bench
Gauntlet time
1.5h
~204s/bench
Speed
14.1
tok/s
Steps/bench
2.3
decisiveness
$ / run
—
OpenRouter-equiv
Overall rank
#24 of 26Agentic
#21 of 22Coding
#20 of 22Speed (tok/s)
#16 of 22Decisive
#21 of 22Strongest
Coding · 67%Weakest
PHP · 20%Where it fits · live, recomputed from current results
#24 of 26 overall; strongest on Coding (67%); weakest on PHP (20%); #16/22 for speed (14.1 tok/s). (auto-generated from current results)
Verdict history · snapshotted daily when it changes
2026-07-08#24 of 26 overall; strongest on Coding (67%); weakest on PHP (20%); #16/22 for speed (14.1 tok/s). (auto-generated from current results)
2026-07-07#25 of 28 overall; strongest on Coding (67%); weakest on SupportFlow (0%); #24/25 on the supportflow gauntlet; #17/26 for speed (14.0 tok/s). (auto-generated from current results)
2026-07-06#25 of 28 overall; strongest on Coding (67%); weakest on SupportFlow (0%); #23/24 on the supportflow gauntlet; #16/26 for speed (14.0 tok/s). (auto-generated from current results)
2026-07-03#25 of 28 overall; strongest on Coding (67%); weakest on PHP (20%); #16/26 for speed (14.1 tok/s). (auto-generated from current results)
2026-07-02#28 of 30 overall; strongest on SupportFlow (0%); weakest on SupportFlow (0%); #23/26 on the supportflow gauntlet. (auto-generated from current results)
2026-07-01#26 of 28 overall; strongest on SupportFlow (0%); weakest on SupportFlow (0%); #20/23 on the supportflow gauntlet. (auto-generated from current results)
2026-06-29#25 of 28 overall; strongest on SupportFlow (0%); weakest on SupportFlow (0%); #18/21 on the supportflow gauntlet. (auto-generated from current results)
Domain breakdown
Gate
36%
4/11
Coding
67%
4/6
Agentic
46%
5/11
C#
33%
5/15
PHP
20%
1/5
Benches run
green = passed · red = failed · hover for harness & time
Gate (11)
Coding (6)
Agentic (11)
C# (15)
PHP (5)
History (monthly pass-rate)
2026-07
47%
54/116
Availability record
When this model appeared in our catalogue census.
No census snapshots recorded yet.