← Models
llama-3.3-70b
evaluatedVendor: MetaQuant: Q8Type: local
Capability starfish
green = capability (pass-rate) · blue = operational (vs roster) · dashed = roster average
Overall
mean domain rate
Solved
39/53
latest per bench
Gauntlet time
14m
~32s/bench
Speed
2.8
tok/s
Steps/bench
1.6
decisiveness
$ / run
—
OpenRouter-equiv
Overall rank
#8 of 26Agentic
#3 of 22Coding
#5 of 22Speed (tok/s)
#22 of 22Decisive
#8 of 22Strongest
Coding · 100%Weakest
Gate · 85%Where it fits · live, recomputed from current results
#8 of 26 overall; strongest on Coding (100%); weakest on Gate (85%); #22/22 for speed (2.8 tok/s). (auto-generated from current results)
Verdict history · snapshotted daily when it changes
2026-07-09#8 of 26 overall; strongest on Coding (100%); weakest on Gate (85%); #22/22 for speed (2.8 tok/s). (auto-generated from current results)
2026-07-08#9 of 26 overall; strongest on Coding (100%); weakest on Gate (85%); #22/22 for speed (2.8 tok/s). (auto-generated from current results)
2026-07-06#4 of 28 overall; strongest on Coding (100%); weakest on Gate (85%); #25/26 for speed (2.8 tok/s). (auto-generated from current results)
2026-07-04#3 of 28 overall; strongest on Coding (100%); weakest on Gate (85%); #25/26 for speed (2.8 tok/s). (auto-generated from current results)
2026-07-03#4 of 28 overall; strongest on Coding (100%); weakest on Gate (85%); #25/26 for speed (2.8 tok/s). (auto-generated from current results)
2026-07-02#2 of 30 overall; strongest on Coding (100%); weakest on Gate (85%); #28/29 for speed (2.8 tok/s). (auto-generated from current results)
2026-06-29#2 of 28 overall; strongest on Coding (100%); weakest on Gate (85%); #26/27 for speed (2.8 tok/s). (auto-generated from current results)
Domain breakdown
Gate
85%
11/13
Coding
100%
6/6
Agentic
100%
6/6
C#
85%
11/13
PHP
100%
5/5
Benches run
green = passed · red = failed · hover for harness & time
Gate (13)
Coding (6)
Agentic (6)
C# (13)
PHP (5)
SWE (10)
History (monthly pass-rate)
2026-06
85%
101/119
Availability record
When this model appeared in our catalogue census.
2026-06-25 · openrouter2026-06-25 · openrouter