Skip to content
Open navigation
OpenOpen workspace

Stack Overflow Developer Survey 2026

LLM vendors: Used in agents by employment status

In the Employed group, 82% selected “Anthropic”; in the I prefer not to say group, 58% did. Of the 12 answers tested, “Anthropic” differs most by employment status, a small association: χ²(5, N = 6827) = 162.64, Holm-adjusted p < .001, V = .15. Respondents could select more than one answer, so the percentages add to more than 100%. These are unweighted respondents, not population estimates.

Published by crosstabs.com · Last updated ·
How results are checkedNamed statistical review pending. Selected automated reference cases are published on the validation page.

The table

The table

Question asked: Which LLM model vendors for AI tools have you used for work in the past year? Which would you like to use next year? Select all that apply. Respondents could select more than one answer. Shown by employment status, as the percentage of each group's respondents who selected each answer, with the number of respondents beneath. A respondent is counted in every row they selected, so a column adds to more than 100%.

LLM vendors: Used in agents by employment status: percentage of each group's respondents who selected each answer, and respondent counts
LLM vendors: Used in agentsEmployed (A)I prefer not to say (B)Independent contractor, freelancer, or self-employed (C)Not employed (D)Retired (E)Student (F)All
Anthropic81.7%BCDEF4,12758.0%2976.8%BDFe71065.6%20860.9%3961.2%25778.7%5,370
OpenAI55.3%2,79356.0%2857.0%52755.8%17740.6%2658.1%24455.6%3,795
Google AI26.3%1,33044.0%a2232.6%A30140.4%A12840.6%2646.0%AC19329.3%2,000
Microsoft15.5%78532.0%ACDFe1612.8%11811.7%3710.9%712.9%5414.9%1,017
DeepSeek8.6%43440.0%ACEF2015.3%A14123.3%AC7410.9%718.3%A7711.0%753
Alibaba Qwen7.4%37622.0%A1112.0%A11111.7%a3712.5%810.7%458.6%588
Mistral AI3.9%19920.0%ADEFc108.1%A755.7%181.6%16.7%284.8%331
Moonshot AI3.3%16520.0%ACDEF107.4%A686.9%A221.6%16.2%A264.3%292
Other, please specify3.2%16214.0%AF76.3%A585.4%176.3%43.3%143.8%262
xAI2.8%14314.0%A76.1%A566.3%A204.7%35.0%213.7%250
Zhipu AI2.0%9920.0%ACDEF105.2%A483.8%121.6%15.5%A232.8%193
Meta AI1.7%8520.0%ACDFe105.0%A465.4%A173.1%26.2%A262.7%186
Databricks (DBRX)1.0%4814.0%ACDF72.1%A192.2%71.6%11.0%41.3%86
Snowflake0.8%4116.0%ACDF82.1%A191.9%60.0%01.2%51.2%79
Cohere0.6%2914.0%ACDF72.1%A192.5%A80.0%01.4%61.0%69
Voyage AI (MongoDB)0.3%1414.0%ACDF72.2%A202.5%A81.6%10.5%20.8%52
Cascade0.3%1712.0%ACDF61.9%A182.2%A70.0%00.7%30.7%51
Black Forest Labs0.3%1416.0%ACDF81.8%A171.9%A60.0%00.7%30.7%48
01.AI0.2%1012.0%ACDF61.7%A161.6%A51.6%10.7%30.6%41
Luel0.2%1012.0%ACDF61.5%A142.2%A70.0%01.0%A40.6%41
Stability AI0.2%1014.0%ACDF71.5%A141.9%A61.6%10.5%20.6%40
Guide Labs0.1%714.0%ACDF71.5%A141.9%A60.0%00.7%a30.5%37
Reka0.2%814.0%ACDF71.3%A121.6%A50.0%00.7%30.5%35
Respondents5,05250924317644206,827

A letter beside a percentage names the columns it is significantly higher than (uppercase 95%, lowercase 90%; pairwise column tests, Bonferroni-corrected). 24,076 of 30,903 respondents have no answer to one of the two questions (not asked or not answered) and are left out.

Analysis

Results

Chi-square test of “Anthropic”

χ² = 162.64

df = 5, Holm-adjusted p < .001

Effect size

Cramér's V = 0.154

a small association

The strongest answer, APA-style: χ²(5, N = 6827) = 162.64, Holm-adjusted p < .001, V = .15. Each answer is tested on its own, as the respondents who selected it against those who did not, across the 6 groups (N = 6,827). The p-values are Holm-adjusted for the 12 answers tested; 11 of them differ by employment status at the 0.05 level. 11 answers are not tested: too few respondents selected them for the test to be reliable (an expected count below 1, or more than 20% of the expected counts below 5).

LLM vendors: Used in agents by employment status: the test of each answer, selected against not selected, by employment status
AnswerSelectedχ² (df = 5)Adjusted pCramér's V
Anthropic78.7%162.64< .0010.154
OpenAI55.6%7.86.1640.034
Google AI29.3%110.55< .0010.127
Microsoft14.9%21.24.0010.056
DeepSeek11.0%162.05< .0010.154
Alibaba Qwen8.6%41.10< .0010.078
Mistral AI4.8%60.31< .0010.094
Moonshot AI4.3%74.64< .0010.105
Other, please specify3.8%37.64< .0010.074
xAI3.7%48.73< .0010.084
Zhipu AI2.8%98.53< .0010.120
Meta AI2.7%122.11< .0010.134
Databricks (DBRX)1.3%Not tested: expected counts too small
Snowflake1.2%Not tested: expected counts too small
Cohere1.0%Not tested: expected counts too small
Voyage AI (MongoDB)0.8%Not tested: expected counts too small
Cascade0.7%Not tested: expected counts too small
Black Forest Labs0.7%Not tested: expected counts too small
01.AI0.6%Not tested: expected counts too small
Luel0.6%Not tested: expected counts too small
Stability AI0.6%Not tested: expected counts too small
Guide Labs0.5%Not tested: expected counts too small
Reka0.5%Not tested: expected counts too small

How to read it

No single chi-square test covers this table. A respondent appears in every row they selected, so the rows are not separate groups of people, and a test of the whole table would count the same person several times. Each answer is tested on its own instead: did the share who selected it differ between the groups? Testing 12 answers gives 12 chances of a false alarm, so the p-values are Holm-adjusted to keep the chance of any false alarm in this table at 0.05.

For “Anthropic” the test rejects independence at the 0.05 level (Holm-adjusted p < .001): differences this large between the groups are unlikely if selecting it did not depend on employment status. Cramér's V of 0.154 is in the small range, which is real but modest: knowing someone's group tells you only a little about their answer.

With 6,827 respondents, even small differences reach statistical significance. Read the percentages and the effect size before the p-value.

Limits

  • The survey has no weights. These tables are unweighted counts of the people who answered, so they describe the respondents, not developers in general.
  • The test treats respondents as a simple random sample. They are people who chose to answer a survey Stack Overflow promoted, not a random sample of developers, so a result holds for these respondents and may not hold beyond them.
  • The adjustment covers the answers of this question by this group only, not the other tables in the library. The headline answer is the strongest of those tested, so its effect size is larger than a typical answer's.
  • A difference between groups is not evidence that group membership causes the answer.

Next steps

Work with this table

Open the test of “Anthropic” in the calculator

The link fills in the respondents who selected this answer and those who did not, by group. The whole table does not open there: its rows overlap, and the calculator's test needs rows that are separate people.

Open this question in the full file

The workspace opens all 30,903 respondents in your browser with this question selected. The survey has no weights, so the table there is unweighted too. Change the group, add a filter or export from there.

Open in the workspace →

Run this on your own data — free, no signup

Upload a CSV, Excel, SPSS, Stata, SAS, R, Parquet or JSON file. Everything runs in your browser; your file never leaves your device.

Open the workspace →

The same question by other groups

Other questions that differ by employment status

Source

Stack Overflow. 2026 Developer Survey, https://survey.stackoverflow.co/2026, licensed under the ODbL 1.0. Tabulated by crosstabs.com. Variables LLMUsedInAgents (LLM vendors: Used in agents) and Employment. Stack Overflow licenses the survey data under the Open Database License (ODbL) 1.0 and its contents under the Database Contents License (DbCL) 1.0: credit Stack Overflow, keep the licence notice, and share any database made from the data under the ODbL. These tables and their downloads are offered on those terms.

  1. [1]Stack Overflow: 2026 Developer Survey public results (survey_results_public.csv)survey.stackoverflow.co
  2. [2]Stack Overflow: Open Database License (ODbL) 1.0opendatacommons.org