Gemini 3.1 Pro Preview (02/26)

Gemini 3.1 Pro Preview (02/26) is a model from Google, released February 19, 2026. Its best result is #2 of 107 on MedCode.

Release Date: Feb 19, 2026

Developer Google  🇺🇸
Context Window 1M
Max Output Tokens 66k
Token Costs (in/out) $2.00/12.00
Weights Private
Input Modalities

Accuracy

49.04 %

Avg. Cost (In/Out)

$ 2.00 / $ 12.00

Latency

9 min 37 s

Vals Index
BenchmarksAccuracyRankings

27.38%

±1.42
25/27

17.31%

±3.93
55/79

52.62%

±2.97
54/76

42.98%

±1.21
60/81

66.73%

±1.50
22/28

20.67%

±2.81
74/98

59.06%

±2.00
2/107

76.11%

±1.92
77/109

69.40%

±0.91
6/98

20.72%

±2.73
28/33

26.00%

±4.41
40/53

83.18%

±1.95
23/29

0.00%

±0.00
36/37

9.29%

±2.09
75/89

72.88%

±0.86
58/145

6.69%

±2.40
26/26

32.03%

±4.34
88/132

60.00%

±2.94
31/33

95.45%

±1.05
1/138

51.83%

±2.59
25/46

88.48%

±0.93
6/143

87.40%

±0.33
5/151

90.99%

±0.28
4/138

88.21%

±0.78
10/93

0.00%

±0.00
50/66

78.80%

±1.83
29/88

2.52%

±0.51
61/68

1.43%

±1.43
40/47
Proprietary BenchmarksAcademic BenchmarksIndustry Partners
Vals
Default Provider : Google
Temperature: Default
Top P: Default
Top K: Default
Max Output Tokens: 65,536
Reasoning Effort: high

Compared with

Updates

Feb 20, 2026

We evaluated Gemini 3.1 Pro Preview (02/26) across our full benchmark suite. Here are the key takeaways:

One notable metric to call out here is the model achieves this performance at a lower cost than models like Claude Opus 4.6, Claude Sonnet 4.6, GPT 5.2 and O3.

Evaluations were run with a temperature of 1.0 and a “high” thinking level, via the official Google API.

Congratulations to the Google team on another outstanding model!

Feb 19, 2026

We evaluated Gemini 3.1 Pro Preview (02/26) on our Vals Index. Here are the key takeaways:

We are evaluating Gemini 3.1 Pro Preview (02/26) on our full suite of benchmarks and will be sharing updates soon!