← Models
qwen3.5-0.8b
evaluatedVendor: AlibabaQuant: Q8Type: local
Capability starfish
green = capability (pass-rate) · blue = operational (vs roster) · dashed = roster average
Overall
mean domain rate
Solved
20/54
latest per bench
Gauntlet time
1m
~2s/bench
Speed
84.7
tok/s
Steps/bench
2.3
decisiveness
$ / run
—
OpenRouter-equiv
Overall rank
#25 of 26Agentic
#18 of 22Coding
#22 of 22Speed (tok/s)
#2 of 22Decisive
#20 of 22Strongest
Agentic · 79%Weakest
Coding · 0%Where it fits · live, recomputed from current results
#25 of 26 overall; strongest on Agentic (79%); weakest on Coding (0%); #2/22 for speed (84.7 tok/s). (auto-generated from current results)
Verdict history · snapshotted daily when it changes
2026-07-08#25 of 26 overall; strongest on Agentic (79%); weakest on Coding (0%); #2/22 for speed (84.7 tok/s). (auto-generated from current results)
2026-07-07#26 of 28 overall; strongest on Agentic (79%); weakest on Coding (0%); #22/25 on the supportflow gauntlet; #2/26 for speed (94.1 tok/s). (auto-generated from current results)
2026-07-06#26 of 28 overall; strongest on Agentic (79%); weakest on Coding (0%); #21/24 on the supportflow gauntlet; #2/26 for speed (94.1 tok/s). (auto-generated from current results)
2026-07-05#26 of 28 overall; strongest on Agentic (79%); weakest on Coding (0%); #18/23 on the supportflow gauntlet; #2/26 for speed (79.6 tok/s). (auto-generated from current results)
2026-07-04#26 of 28 overall; strongest on Agentic (79%); weakest on Coding (0%); #19/23 on the supportflow gauntlet; #2/26 for speed (79.6 tok/s). (auto-generated from current results)
2026-07-03#26 of 28 overall; strongest on Agentic (79%); weakest on Coding (0%); #18/22 on the supportflow gauntlet; #2/26 for speed (79.6 tok/s). (auto-generated from current results)
2026-07-02#26 of 30 overall; strongest on Agentic (79%); weakest on Coding (0%); #20/26 on the supportflow gauntlet; #2/29 for speed (67.9 tok/s). (auto-generated from current results)
2026-07-01#23 of 28 overall; strongest on Agentic (67%); weakest on Coding (0%); #2/27 for speed (118.3 tok/s). (auto-generated from current results)
2026-06-29#21 of 28 overall; strongest on Agentic (67%); weakest on Coding (0%); #2/27 for speed (118.3 tok/s). (auto-generated from current results)
Domain breakdown
Gate
64%
9/14
Coding
0%
0/6
Agentic
79%
11/14
C#
0%
0/15
PHP
0%
0/5
Benches run
green = passed · red = failed · hover for harness & time
Gate (14)
Coding (6)
Agentic (14)
C# (15)
PHP (5)
History (monthly pass-rate)
2026-06
49%
110/224
2026-07
73%
480/656
Availability record
When this model appeared in our catalogue census.
No census snapshots recorded yet.