BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases12

EnterpriseOps-Gym leaderboard

As of 9 Nov 2026, the top score on EnterpriseOps-Gym is 51.1% by Claude Fable 5 (Anthropic), ahead of Gemini 3.7 Flash at 50.4%, across 51 model configurations with a published result (Artificial Analysis, result dated 9 Jun 2026). Enterprise operations tasks in a simulated company: tickets, approvals and systems a model has to drive to finish a job. EnterpriseOps-Gym is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 9 Nov 2026)

  • Claude Fable 5 holds the top EnterpriseOps-Gym score at 51.1% (Artificial Analysis, 9 Jun 2026).
  • Gemini 3.7 Flash is second on EnterpriseOps-Gym at 50.4% (Artificial Analysis, 13 Aug 2026).
  • EnterpriseOps-Gym: Enterprise operations tasks in a simulated company: tickets, approvals and systems a model has to drive to finish a job (Artificial Analysis).
  • 51 model configurations have a published EnterpriseOps-Gym result on BenchLeader (BenchLeader, 9 Nov 2026).
  • EnterpriseOps-Gym does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Enterprise operations tasks in a simulated company: tickets, approvals and systems a model has to drive to finish a job.

Published by
Artificial Analysis
Category
Agents & tools
Index weight
Reference only
Models
51
Data as of
9 Nov 2026

Source: Artificial Analysis (artificialanalysis.ai). Data taken from the public leaderboard.

Top 15
  1. 1Claude Fable 5 (max)51.1%
  2. 2Gemini 3.7 Flash (medium)50.4%
  3. 3Gemini 3.5 Flash (high)50.1%
  4. 4DeepSeek V4 Pro (max)49.6%
  5. 5Grok 4.6 (high)48.3%
  6. 6Qwen3.8 Max (max)47.6%
  7. 7Claude Opus 5 (max)47.5%
  8. 8Qwen3.8 2.4T A95B47.4%
  9. 9Muse Spark 1.2 (xhigh)47.3%
  10. 10Step 5 Preview47.2%
  11. 11Deepseek v4 Flash Vision (max)47.1%
  12. 12Inkling 2 Small46.8%
  13. 13GPT-5.5 (xhigh)46.6%
  14. 14Kimi K3 (max)45.3%
  15. 15Qwen3.7 (max)45.0%
51 of 51
#
1Claude Fable 5max51.1%65.32026-06-09
2Gemini 3.7 Flashmedium50.4%60.22026-08-13
3Gemini 3.5 Flashhigh50.1%62.52026-05-19
4DeepSeek V4 Promax49.6%56.92026-08-13
5Grok 4.6high48.3%60.52026-08-12
6Qwen3.8 Maxmax47.6%58.92026-09-02
7Claude Opus 5max47.5%65.12026-07-24
8Qwen3.8 2.4T A95B47.4%51.62026-08-12
9Muse Spark 1.2xhigh47.3%61.72026-08-05
10Step 5 Preview47.2%54.62026-09-18
11Deepseek v4 Flash Visionmax47.1%–2026-08-21
12NewInkling 2 Small46.8%58.32026-11-05
13GPT-5.5xhigh46.6%64.32026-04-23
14Kimi K3max45.3%62.02026-07-16
15Qwen3.7max45.0%60.32026-05-19
16Claude Sonnet 5max44.7%53.62026-06-30
17Qwen3.8 27Bxhigh44.2%52.42026-08-14
18GPT-5.5medium44.1%60.92026-04-23
19Claude Opus 4.8max44.0%60.22026-05-28
20Claude Sonnet 5xhigh43.9%55.52026-06-30
21GPT-5.6 Solmax42.9%65.42026-07-09
22GLM 5.2max42.7%59.42026-06-16
23Inkling Small42.7%52.12026-07-30
24Gemini 3.5 Flash Lite42.4%51.52026-07-21
25Gemini 3.1 Pro42.2%60.62026-02-19
26Grok 4.5high40.8%56.82026-07-08
27GPT-5.6 Lunamax40.8%56.22026-07-09
28Qwen3.7 Plus40.6%57.72026-06-01
29DeepSeek V4 Pro 0424max40.4%–2026-04-24
30Kimi K2.7 Code40.2%54.22026-06-12
31DeepSeek V4 Flash 0420max39.6%–2026-04-24
32Kimi K2.6thinking38.5%59.82026-04-20
33GPT-5.6 Terramax38.5%60.52026-07-09
34Inklingxhigh38.0%50.12026-07-15
35GLM 5.3max36.4%58.92026-08-18
36Muse Glimmerhigh34.7%46.62026-08-10
37GPT-5.4 Minixhigh34.6%56.02026-03-17
38Mistral Medium 3.533.7%46.62026-04-29
39GLM 5.3 Flash33.2%56.82026-08-26
40MiniMax M332.1%56.32026-06-01
41GPT-5.4 Nanoxhigh31.5%54.02026-03-17
42Qwen3.5 397B A17Bthinking30.9%56.52026-02-16
43Claude Haiku 4.5thinking29.4%45.32025-10-15
44Step 3.7 Flash29.1%53.12026-05-29
45Nemotron 3 Ultra 550B A55Bthinking28.9%49.22026-06-04
46Gemma 4 31Bthinking28.4%46.62026-04-02
47Gemini 3.1 Flash Lite28.0%52.42026-03-03
48Grok 4.3high26.3%56.62026-04-30
49gpt-oss-120bhigh25.5%48.32025-08-05
50Nemotron 3 Super 120B A12Bthinking23.6%51.22026-03-11
51Nemotron 3.5 Lightning18.4%41.52026-08-11

More agents & tools benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on EnterpriseOps-Gym

Cite as: BenchLeader, “EnterpriseOps-Gym leaderboard”, data as of 9 Nov 2026, https://www.benchleader.com/benchmarks/aa_enterprise_ops_gym.

EnterpriseOps-Gym: questions

What does EnterpriseOps-Gym measure?
Enterprise operations tasks in a simulated company: tickets, approvals and systems a model has to drive to finish a job. Scores are reported in percent of tasks solved; higher is better.
What is the top score on EnterpriseOps-Gym?
As of 9 Nov 2026, the highest score on EnterpriseOps-Gym is 51.1% by Claude Fable 5 (result dated 9 Jun 2026), ahead of Gemini 3.7 Flash at 50.4%. Scores come from Artificial Analysis and are re-read daily.
Which AI model leads EnterpriseOps-Gym?
Claude Fable 5 leads EnterpriseOps-Gym with 51.1% as of 9 Nov 2026 (result dated 9 Jun 2026), ahead of Gemini 3.7 Flash at 50.4%.
How many models have EnterpriseOps-Gym results?
51 model configurations have a EnterpriseOps-Gym result on BenchLeader, all taken from Artificial Analysis.
Who runs EnterpriseOps-Gym and how often is it updated?
EnterpriseOps-Gym is published by Artificial Analysis. BenchLeader re-reads the published results once a day and records the date each result was published.
Does EnterpriseOps-Gym count toward the BenchLeader Index?
No. EnterpriseOps-Gym is shown for reference but left out of the composite index.