devstral-small-2-2512

evaluatedCoding specialist

Vendor: Mistral AIParams: 24BQuant: Q8Context: 128kLicense: Apache-2.0Type: local

Capability starfish

green = capability (pass-rate) · blue = operational (vs roster) · dashed = roster average

Overall

mean domain rate

Solved

43/62

latest per bench

Gauntlet time

~9s/bench

Speed

8.2

tok/s

Steps/bench

1.7

decisiveness

$ / run

$0.0004

OpenRouter-equiv

Overall rank

#17 of 26

Supportflow

#1 of 5

Agentic

#10 of 22

Coding

#13 of 22

Speed (tok/s)

#17 of 22

Decisive

#14 of 22

Cheapest

#8 of 9

Strongest

Gate · 93%

Weakest

SupportFlow · 67%

Where it fits · live, recomputed from current results

#17 of 26 overall; strongest on Gate (93%); weakest on SupportFlow (67%); #1/5 on the supportflow gauntlet; #17/22 for speed (8.2 tok/s); #8/9 cheapest ($0.0004/run). (auto-generated from current results)

curator note: Purpose-built for code edits. Solid on focused fixes but dense (slow tok/s) and pricier equivalent. Good when correctness on a single-file change beats throughput.

codingdenseslow

Verdict history · snapshotted daily when it changes

2026-07-09#17 of 26 overall; strongest on Gate (93%); weakest on SupportFlow (67%); #1/5 on the supportflow gauntlet; #17/22 for speed (8.2 tok/s); #8/9 cheapest ($0.0004/run). (auto-generated from current results)

2026-07-08#17 of 26 overall; strongest on Gate (93%); weakest on SupportFlow (67%); #1/4 on the supportflow gauntlet; #17/22 for speed (8.2 tok/s); #8/9 cheapest ($0.0004/run). (auto-generated from current results)

2026-07-07#17 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (42%); #8/25 on the supportflow gauntlet; #20/26 for speed (7.6 tok/s); #7/9 cheapest ($0.0109/run). (auto-generated from current results)

2026-07-06#16 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (42%); #9/24 on the supportflow gauntlet; #20/26 for speed (7.6 tok/s); #7/9 cheapest ($0.0109/run). (auto-generated from current results)

2026-07-05#16 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (42%); #10/23 on the supportflow gauntlet; #20/26 for speed (7.6 tok/s); #7/9 cheapest ($0.0109/run). (auto-generated from current results)

2026-07-04#16 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (42%); #8/23 on the supportflow gauntlet; #20/26 for speed (7.6 tok/s); #7/9 cheapest ($0.0109/run). (auto-generated from current results)

2026-07-03#16 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (35%); #9/22 on the supportflow gauntlet; #20/26 for speed (7.6 tok/s); #6/9 cheapest ($0.0074/run). (auto-generated from current results)

2026-07-02#19 of 30 overall; strongest on Gate (93%); weakest on PHP (0%); #14/26 on the supportflow gauntlet; #22/29 for speed (7.6 tok/s); #7/9 cheapest ($0.0074/run). (auto-generated from current results)

2026-07-01#11 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (35%); #8/23 on the supportflow gauntlet; #21/27 for speed (7.6 tok/s); #6/9 cheapest ($0.0074/run). (auto-generated from current results)

2026-06-30#11 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (35%); #8/21 on the supportflow gauntlet; #21/27 for speed (7.6 tok/s); #6/9 cheapest ($0.0074/run). (auto-generated from current results)

2026-06-29#10 of 28 overall; strongest on Gate (93%); weakest on SupportFlow (35%); #8/21 on the supportflow gauntlet; #21/27 for speed (7.6 tok/s); #6/9 cheapest ($0.0074/run). (auto-generated from current results)

Domain breakdown

Gate

93%

13/14

Coding

83%

5/6

Agentic

89%

8/9

SupportFlow

67%

2/3

73%

11/15

PHP

80%

4/5

Benches run

green = passed · red = failed · hover for harness & time

Gate (14)

Coding (6)

Agentic (9)

SupportFlow (3)

C# (15)

PHP (5)

SWE (10)

History (monthly pass-rate)

2026-06

82%

302/370

Availability record

When this model appeared in our catalogue census.

No census snapshots recorded yet.