Benchmark Almanac

Spatial Agent Index v2

Claude Opus 5.5 86.0 leads since November 2026
Not in model cards
Version 2

Agents analyze geospatial files and produce numbers, maps, or lists. Axis Spatial averages the scores for each included evaluation.

Axis Spatial allows entries with a score on every included evaluation and at least 75% coverage. Code checks data outputs. A model grades maps.

Made by
Axis Spatial
Released
November 2026
Kind
Apps and MCP servers
Level
University
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
Interaction
Agentic
Environment
APIs and tools
Items from
Another benchmark
Run by
Anyone
In views
Agents
Strongest model tested
Claude Opus 5.5 #5 on the capability index
Last model added
October 2026
Scores refreshed
November 20th, 2026
48
Entries on its own leaderboard, 47 models we track

Claude Opus 5.5 leads with 86.0 Claude Opus 5.5 leads with 86.0

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

  1. 1 Claude Opus 5.5 max 86.0
  2. 2 GPT-6.1 Sol xhigh 82.0
  3. 3 Claude Fable 5.1 medium 81.0
  4. 4 Claude Opus 5 high 80.0
  5. 5 GPT-6 Sol high 80.0
  6. 6 Claude Sonnet 5.5 max 79.0
  7. 7 Claude Haiku 5.5 max 77.0
  8. 8 Claude Opus 4.8 high 76.0
  9. 9 GPT-5.6 Sol high 76.0
  10. 10 DeepSeek V4.1 Flash max 76.0
  11. 11 Kimi K3 Reasoning 75.0
  12. 12 Claude Opus 4.7 high 74.0
  13. 13 GLM-5.2 max 74.0
  14. 14 Claude Sonnet 5 high 74.0
  15. 15 GLM-5.3 max 74.0
  16. 16 GPT-5.6 Luna xhigh 73.0
  17. 17 GPT-6 Luna xhigh 73.0
  18. 18 DeepSeek-V4-Flash high 72.0
  19. 19 DeepSeek-V4-Pro max 72.0
  20. 20 Kimi K2.7 Code Reasoning 72.0
  21. 21 GLM-5.3-Flash max 72.0
  22. 22 Gemini 3.8 Flash low 72.0
  23. 23 MiMo-V2.6 Flash Reasoning 71.0
  24. 24 Kimi K2.6 Reasoning 70.0
  25. 25 GPT-5.6 Terra high 69.0
  26. 26 Qwen 3.8 27B Reasoning 69.0
  27. 27 MiniMax-M3 Reasoning 68.0
  28. 28 Nemotron 3 Ultra high 68.0
  29. 29 Muse Glimmer high 64.0
  30. 30 Mistral Large 4 high 63.0
  31. 31 Qwen 3.6 35B-A3B Reasoning 60.0
  32. 32 Claude Haiku 4.5 Reasoning 58.0
  33. 33 Nemotron 3.5 Super 120B Reasoning 56.0
  34. 34 Gemini 3.1 Flash-Lite high 56.0
  35. 35 Gemma 4 26B A4B Reasoning 56.0
  36. 36 Gemma 4 31B IT Reasoning 55.0
  37. 37 MiniMax-M3.1-Flash max 53.0
  38. 38 Step 5 Preview Reasoning 52.0
  39. 39 Nemotron 3.5 Super VL Reasoning 52.0
  40. 40 gpt-oss-120b high 46.0
  41. 41 glm-4.7-flash Reasoning 43.0
  42. 42 Laguna XS 2.1 Reasoning 37.0
  43. 43 gpt-oss-20b high 33.0
  44. 44 Mistral Large 3 Non-reasoning 33.0
  45. 45 Nemotron 3 Nano Omni 30B-A3B Reasoning 21.0
  46. 46 Nemotron 3.5 Lightning 30B-A3B Reasoning 13.0
  47. 47 Llama 3.3 70B Non-reasoning 8.0
  48. 48 Nemotron 3.5 Bolt E4B Reasoning 4.0

Source: Official Spatial Agent Index v2 leaderboard

Source: Official Spatial Agent Index v2 leaderboard

#EntryScoreSettingDateLinkSource
1 Claude Opus 5.5 86.0 max, Claude Code Sep 2026 axisspatial.com Official leaderboard
2 GPT-6.1 Sol 82.0 xhigh, Codex Sep 2026 axisspatial.com Official leaderboard
3 Claude Fable 5.1 81.0 medium, Claude Code Sep 2026 axisspatial.com Official leaderboard
4 Claude Opus 5 80.0 high, Claude Code Jul 2026 axisspatial.com Official leaderboard
5 GPT-6 Sol 80.0 high, Codex Sep 2026 axisspatial.com Official leaderboard
6 Claude Sonnet 5.5 79.0 max, Claude Code Sep 2026 axisspatial.com Official leaderboard
7 Claude Haiku 5.5 77.0 max, Claude Code Oct 2026 axisspatial.com Official leaderboard
8 Claude Opus 4.8 76.0 high, Claude Code May 2026 axisspatial.com Official leaderboard
9 GPT-5.6 Sol 76.0 high, Codex Jul 2026 axisspatial.com Official leaderboard
10 DeepSeek V4.1 Flash 76.0 max, Codex Sep 2026 axisspatial.com Official leaderboard
11 Kimi K3 75.0 Reasoning, Kimi Code Jul 2026 axisspatial.com Official leaderboard
12 Claude Opus 4.7 74.0 high, Claude Code Apr 2026 axisspatial.com Official leaderboard
13 GLM-5.2 74.0 max, Claude Code Jun 2026 axisspatial.com Official leaderboard
14 Claude Sonnet 5 74.0 high, Claude Code Jun 2026 axisspatial.com Official leaderboard
15 GLM-5.3 74.0 max, Claude Code Aug 2026 axisspatial.com Official leaderboard
16 GPT-5.6 Luna 73.0 xhigh, Codex Jul 2026 axisspatial.com Official leaderboard
17 GPT-6 Luna 73.0 xhigh, Codex Sep 2026 axisspatial.com Official leaderboard
18 DeepSeek-V4-Flash 72.0 high, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
19 DeepSeek-V4-Pro 72.0 max, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
20 Kimi K2.7 Code 72.0 Reasoning, Kimi Code Jun 2026 axisspatial.com Official leaderboard
21 GLM-5.3-Flash 72.0 max, Claude Code Aug 2026 axisspatial.com Official leaderboard
22 Gemini 3.8 Flash 72.0 low, Terminus-2 Sep 2026 axisspatial.com Official leaderboard
23 MiMo-V2.6 Flash 71.0 Reasoning, MiMo Code Sep 2026 axisspatial.com Official leaderboard
24 Kimi K2.6 70.0 Reasoning, Kimi Code Apr 2026 axisspatial.com Official leaderboard
25 GPT-5.6 Terra 69.0 high, Codex Jul 2026 axisspatial.com Official leaderboard
26 Qwen 3.8 27B 69.0 Reasoning, Qwen Code Aug 2026 axisspatial.com Official leaderboard
27 MiniMax-M3 68.0 Reasoning, MiniMax Code Jun 2026 axisspatial.com Official leaderboard
28 Nemotron 3 Ultra 68.0 high, Terminus-2 Jun 2026 axisspatial.com Official leaderboard
29 Muse Glimmer 64.0 high, Terminus-2 Aug 2026 axisspatial.com Official leaderboard
30 Mistral Large 4 63.0 high, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
31 Qwen 3.6 35B-A3B 60.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
32 Claude Haiku 4.5 58.0 Reasoning, Claude Code Oct 2025 axisspatial.com Official leaderboard
33 Nemotron 3.5 Super 120B 56.0 Reasoning, Terminus-2 axisspatial.com Official leaderboard
34 Gemini 3.1 Flash-Lite 56.0 high, Terminus-2 Mar 2026 axisspatial.com Official leaderboard
35 Gemma 4 26B A4B 56.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
36 Gemma 4 31B IT 55.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
37 MiniMax-M3.1-Flash 53.0 max, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
38 Step 5 Preview 52.0 Reasoning, Terminus-2 Sep 2026 axisspatial.com Official leaderboard
39 Nemotron 3.5 Super VL 52.0 Reasoning, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
40 gpt-oss-120b 46.0 high, Terminus-2 Aug 2025 axisspatial.com Official leaderboard
41 glm-4.7-flash 43.0 Reasoning, Terminus-2 Jan 2026 axisspatial.com Official leaderboard
42 Laguna XS 2.1 37.0 Reasoning, Terminus-2 Jul 2026 axisspatial.com Official leaderboard
43 gpt-oss-20b 33.0 high, Terminus-2 Aug 2025 axisspatial.com Official leaderboard
44 Mistral Large 3 33.0 Non-reasoning, Terminus-2 Dec 2025 axisspatial.com Official leaderboard
45 Nemotron 3 Nano Omni 30B-A3B 21.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
46 Nemotron 3.5 Lightning 30B-A3B 13.0 Reasoning, Terminus-2 Aug 2026 axisspatial.com Official leaderboard
47 Llama 3.3 70B 8.0 Non-reasoning, Terminus-2 Dec 2024 axisspatial.com Official leaderboard
48 Nemotron 3.5 Bolt E4B 4.0 Reasoning, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
#ModelScoreSettingDateLinkSource
1 Claude Opus 5.5 86.0 max, Claude Code Sep 2026 axisspatial.com Official leaderboard
2 GPT-6.1 Sol 82.0 xhigh, Codex Sep 2026 axisspatial.com Official leaderboard
3 Claude Fable 5.1 81.0 medium, Claude Code Sep 2026 axisspatial.com Official leaderboard
4 Claude Opus 5 80.0 high, Claude Code Jul 2026 axisspatial.com Official leaderboard
5 GPT-6 Sol 80.0 high, Codex Sep 2026 axisspatial.com Official leaderboard
6 Claude Sonnet 5.5 79.0 max, Claude Code Sep 2026 axisspatial.com Official leaderboard
7 Claude Haiku 5.5 77.0 max, Claude Code Oct 2026 axisspatial.com Official leaderboard
8 Claude Opus 4.8 76.0 high, Claude Code May 2026 axisspatial.com Official leaderboard
9 GPT-5.6 Sol 76.0 high, Codex Jul 2026 axisspatial.com Official leaderboard
10 DeepSeek V4.1 Flash 76.0 max, Codex Sep 2026 axisspatial.com Official leaderboard
11 Kimi K3 75.0 Reasoning, Kimi Code Jul 2026 axisspatial.com Official leaderboard
12 Claude Opus 4.7 74.0 high, Claude Code Apr 2026 axisspatial.com Official leaderboard
13 GLM-5.2 74.0 max, Claude Code Jun 2026 axisspatial.com Official leaderboard
14 Claude Sonnet 5 74.0 high, Claude Code Jun 2026 axisspatial.com Official leaderboard
15 GLM-5.3 74.0 max, Claude Code Aug 2026 axisspatial.com Official leaderboard
16 GPT-5.6 Luna 73.0 xhigh, Codex Jul 2026 axisspatial.com Official leaderboard
17 GPT-6 Luna 73.0 xhigh, Codex Sep 2026 axisspatial.com Official leaderboard
18 DeepSeek-V4-Flash 72.0 high, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
19 DeepSeek-V4-Pro 72.0 max, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
20 Kimi K2.7 Code 72.0 Reasoning, Kimi Code Jun 2026 axisspatial.com Official leaderboard
21 GLM-5.3-Flash 72.0 max, Claude Code Aug 2026 axisspatial.com Official leaderboard
22 Gemini 3.8 Flash 72.0 low, Terminus-2 Sep 2026 axisspatial.com Official leaderboard
23 MiMo-V2.6 Flash 71.0 Reasoning, MiMo Code Sep 2026 axisspatial.com Official leaderboard
24 Kimi K2.6 70.0 Reasoning, Kimi Code Apr 2026 axisspatial.com Official leaderboard
25 GPT-5.6 Terra 69.0 high, Codex Jul 2026 axisspatial.com Official leaderboard
26 Qwen 3.8 27B 69.0 Reasoning, Qwen Code Aug 2026 axisspatial.com Official leaderboard
27 MiniMax-M3 68.0 Reasoning, MiniMax Code Jun 2026 axisspatial.com Official leaderboard
28 Nemotron 3 Ultra 68.0 high, Terminus-2 Jun 2026 axisspatial.com Official leaderboard
29 Muse Glimmer 64.0 high, Terminus-2 Aug 2026 axisspatial.com Official leaderboard
30 Mistral Large 4 63.0 high, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
31 Qwen 3.6 35B-A3B 60.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
32 Claude Haiku 4.5 58.0 Reasoning, Claude Code Oct 2025 axisspatial.com Official leaderboard
33 Gemini 3.1 Flash-Lite 56.0 high, Terminus-2 Mar 2026 axisspatial.com Official leaderboard
34 Gemma 4 26B A4B 56.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
35 Gemma 4 31B IT 55.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
36 MiniMax-M3.1-Flash 53.0 max, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
37 Step 5 Preview 52.0 Reasoning, Terminus-2 Sep 2026 axisspatial.com Official leaderboard
38 Nemotron 3.5 Super VL 52.0 Reasoning, Terminus-2 Oct 2026 axisspatial.com Official leaderboard
39 gpt-oss-120b 46.0 high, Terminus-2 Aug 2025 axisspatial.com Official leaderboard
40 glm-4.7-flash 43.0 Reasoning, Terminus-2 Jan 2026 axisspatial.com Official leaderboard
41 Laguna XS 2.1 37.0 Reasoning, Terminus-2 Jul 2026 axisspatial.com Official leaderboard
42 gpt-oss-20b 33.0 high, Terminus-2 Aug 2025 axisspatial.com Official leaderboard
43 Mistral Large 3 33.0 Non-reasoning, Terminus-2 Dec 2025 axisspatial.com Official leaderboard
44 Nemotron 3 Nano Omni 30B-A3B 21.0 Reasoning, Terminus-2 Apr 2026 axisspatial.com Official leaderboard
45 Nemotron 3.5 Lightning 30B-A3B 13.0 Reasoning, Terminus-2 Aug 2026 axisspatial.com Official leaderboard
46 Llama 3.3 70B 8.0 Non-reasoning, Terminus-2 Dec 2024 axisspatial.com Official leaderboard
47 Nemotron 3.5 Bolt E4B 4.0 Reasoning, Terminus-2 Oct 2026 axisspatial.com Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Earth-Bench V2, GeoBenchX and GS-QA2. Claude Opus 5.5 leads 3 of them, and GPT-6.1 Sol leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Earth-Bench V2

47 models we track have a score here.

  1. 1 Claude Opus 5.5 80.0
  2. 2 GLM-5.3 ↑13 77.0
  3. 3 DeepSeek V4.1 Flash ↑7 77.0
  4. 4 Kimi K3 ↑7 75.0
  5. 5 GPT-6 Sol 75.0
  6. 6 Claude Opus 5 ↓2 74.0
  7. 7 DeepSeek-V4-Flash ↑11 73.0
  8. 8 Kimi K2.7 Code ↑12 73.0
  9. 9 Claude Fable 5.1 ↓6 73.0
  10. 10 GPT-6.1 Sol ↓8 72.0

The best score, 86.0, costs $1.14 a task

Runs on the cost frontier score higher than any run that's cheaper. GPT-6.1 Sol at medium comes within 4 of the best for $0.05 a task, 25 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Anthropic OpenAI Cost frontier One model at each effort level
0.0 20.0 40.0 60.0 80.0 100.0 $0.001 $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 5.5, max Claude Opus 5.5, xhigh GPT-6.1 Sol, medium Claude Haiku 5.5, high Claude Haiku 5.5, medium GPT-6 Luna, medium 0.0 20.0 40.0 60.0 80.0 100.0 $0.001 $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 5.5, max Claude Haiku 5.5, high

Source: Official Spatial Agent Index v2 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 max Claude Code 86.0 $1.14 29,000
Claude Opus 5.5 xhigh Claude Code 85.0 $0.25 5,000
Claude Opus 5.5 high Claude Code 82.0 $0.13 2,000
GPT-6.1 Sol xhigh Codex 82.0 $0.09 3,000
GPT-6.1 Sol medium Codex 82.0 $0.05 1,000
Claude Fable 5.1 medium Claude Code 81.0 $0.33 2,000
Claude Opus 5 high Claude Code 80.0 $0.32 4,000
Claude Fable 5.1 high Claude Code 80.0 $0.44 3,000
Claude Opus 5.5 medium Claude Code 80.0 $0.11 2,000
GPT-6 Sol high Codex 80.0 $0.08 3,000
GPT-6.1 Sol high Codex 80.0 $0.06 2,000
Claude Sonnet 5.5 max Claude Code 79.0 $0.98 52,000
Claude Sonnet 5.5 xhigh Claude Code 79.0 $0.11 4,000
Claude Sonnet 5.5 high Claude Code 78.0 $0.06 2,000
Claude Haiku 5.5 max Claude Code 77.0 $0.09 45,000
Claude Opus 4.8 high Claude Code 76.0 $0.26 4,000
GPT-5.6 Sol high Codex 76.0 $0.21 4,000
DeepSeek V4.1 Flash max Codex 76.0 $0.08 41,000
DeepSeek V4.1 Flash max Claude Code 76.0 $0.20 48,000
Claude Haiku 5.5 xhigh Claude Code 76.0 $0.02 12,000
Claude Haiku 5.5 high Claude Code 76.0 $0.009 6,000
Kimi K3 Reasoning Kimi Code 75.0 $0.32 9,000
Claude Opus 4.7 high Claude Code 74.0 $0.34 4,000
GLM-5.2 max Claude Code 74.0 $0.34 28,000
Claude Sonnet 5 high Claude Code 74.0 $0.20 6,000
GLM-5.3 max Claude Code 74.0 $0.48 15,000
Claude Opus 5.5 low Claude Code 74.0 $0.08 1,000
GPT-5.6 Luna xhigh Codex 73.0 $0.02 7,000
Kimi K3 Reasoning Terminus-2 73.0 $0.23 6,000
GPT-6 Luna xhigh Codex 73.0 $0.007 6,000
DeepSeek-V4-Flash high Terminus-2 72.0 $0.07 31,000
DeepSeek-V4-Pro max Terminus-2 72.0 $0.35 24,000
Kimi K2.7 Code Reasoning Kimi Code 72.0 $0.18 11,000
Kimi K2.7 Code Reasoning Terminus-2 72.0 $0.11 12,000
GLM-5.2 max Terminus-2 72.0 $0.32 30,000
GPT-5.6 Luna high Codex 72.0 $0.02 5,000
GLM-5.3 max Terminus-2 72.0 $0.38 19,000
GLM-5.3-Flash max Claude Code 72.0 $0.09 29,000
Gemini 3.8 Flash low Terminus-2 72.0 $0.11 8,000
Claude Haiku 5.5 medium Claude Code 72.0 $0.006 4,000
GLM-5.2 high Terminus-2 71.0 $0.22 22,000
MiMo-V2.6 Flash Reasoning MiMo Code 71.0 $0.05 32,000
GPT-6 Luna high Codex 71.0 $0.004 3,000
Claude Sonnet 5.5 medium Claude Code 71.0 $0.05 1,000
Kimi K2.6 Reasoning Kimi Code 70.0 $0.27 18,000
Kimi K2.6 Reasoning Terminus-2 70.0 $0.09 12,000
DeepSeek-V4-Pro high Terminus-2 70.0 $0.30 20,000
Claude Haiku 5.5 low Claude Code 70.0 $0.005 3,000
DeepSeek-V4-Flash max Terminus-2 69.0 $0.06 29,000
GPT-5.6 Terra high Codex 69.0 $0.10 3,000
Qwen 3.8 27B Reasoning Qwen Code 69.0 $0.22 28,000
Claude Sonnet 5.5 low Claude Code 69.0 $0.04 1,000
MiniMax-M3 Reasoning MiniMax Code 68.0 $0.15 30,000
Nemotron 3 Ultra high Terminus-2 68.0 $0.08 9,000
GLM-5.3 low Terminus-2 68.0 $0.27 10,000
Qwen 3.8 27B xhigh Terminus-2 67.0 $0.12 22,000
DeepSeek-V4-Flash high Codex 66.0 $0.09 19,000
GPT-5.6 Luna medium Codex 66.0 $0.009 3,000
Muse Glimmer high Terminus-2 64.0 $0.08 11,000
GLM-5.2 Non-reasoning Terminus-2 63.0 $0.10 6,000
GLM-5.3-Flash low Terminus-2 63.0 $0.05 15,000
GLM-5.3-Flash max Terminus-2 63.0 $0.09 30,000
Mistral Large 4 high Terminus-2 63.0 $0.10 14,000
GPT-5.6 Luna low Codex 61.0 $0.006 2,000
Qwen 3.6 35B-A3B Reasoning Terminus-2 60.0 $0.07 16,000
DeepSeek-V4-Flash high Claude Code 59.0 $0.08 19,000
Claude Haiku 4.5 Reasoning Claude Code 58.0 $0.09 7,000
Qwen 3.6 35B-A3B Reasoning Qwen Code 57.0 18,000
GPT-6 Luna medium Codex 57.0 $0.003 1,000
Gemini 3.1 Flash-Lite high Terminus-2 56.0 $0.03 16,000
Gemma 4 26B A4B Reasoning Terminus-2 56.0 $0.02 21,000
Gemma 4 31B IT Reasoning Terminus-2 55.0 $0.02 8,000
MiniMax-M3.1-Flash max Terminus-2 53.0 $0.02 19,000
Step 5 Preview Reasoning Terminus-2 52.0 $0.13 24,000
Nemotron 3.5 Super VL Reasoning Terminus-2 52.0 $0.01 7,000
GPT-6 Luna low Codex 51.0 $0.003 1,000
Gemini 3.1 Flash-Lite low Terminus-2 48.0 $0.02 8,000
gpt-oss-120b high Terminus-2 46.0 $0.03 15,000
gpt-oss-120b high Codex 43.0 $0.26 27,000
glm-4.7-flash Reasoning Terminus-2 43.0 $0.04 14,000
glm-4.7-flash Reasoning Claude Code 41.0 $0.27 15,000
Laguna XS 2.1 Reasoning Terminus-2 37.0 $0.05 24,000
gpt-oss-20b high Terminus-2 33.0 $0.02 40,000
Mistral Large 3 Non-reasoning Terminus-2 33.0 $0.03 4,000
Gemini 3.1 Flash-Lite minimal Terminus-2 32.0 $0.007 2,000
gpt-oss-120b low Terminus-2 31.0 $0.005 2,000
Nemotron 3 Nano Omni 30B-A3B Reasoning Terminus-2 21.0 21,000
Nemotron 3.5 Lightning 30B-A3B Reasoning Terminus-2 13.0 $0.01 7,000
gpt-oss-20b high Codex 9.0 $0.09 23,000
Llama 3.3 70B Non-reasoning Terminus-2 8.0 $0.02 2,000
Nemotron 3.5 Bolt E4B Reasoning Terminus-2 4.0 14,000

The best score, 86.0, costs $1.14 a task

Runs on the cost frontier score higher than any run that's cheaper. GPT-6.1 Sol at medium comes within 4 of the best for $0.05 a task, 25 times cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Anthropic OpenAI Cost frontier One model at each effort level
0.0 20.0 40.0 60.0 80.0 100.0 $0.001 $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 5.5, max Claude Opus 5.5, xhigh GPT-6.1 Sol, medium Claude Haiku 5.5, high Claude Haiku 5.5, medium GPT-6 Luna, medium 0.0 20.0 40.0 60.0 80.0 100.0 $0.001 $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 5.5, max Claude Haiku 5.5, high

Source: Official Spatial Agent Index v2 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 max Claude Code 86.0 $1.14 29,000
Claude Opus 5.5 xhigh Claude Code 85.0 $0.25 5,000
Claude Opus 5.5 high Claude Code 82.0 $0.13 2,000
GPT-6.1 Sol xhigh Codex 82.0 $0.09 3,000
GPT-6.1 Sol medium Codex 82.0 $0.05 1,000
Claude Fable 5.1 medium Claude Code 81.0 $0.33 2,000
Claude Opus 5 high Claude Code 80.0 $0.32 4,000
Claude Fable 5.1 high Claude Code 80.0 $0.44 3,000
Claude Opus 5.5 medium Claude Code 80.0 $0.11 2,000
GPT-6 Sol high Codex 80.0 $0.08 3,000
GPT-6.1 Sol high Codex 80.0 $0.06 2,000
Claude Sonnet 5.5 max Claude Code 79.0 $0.98 52,000
Claude Sonnet 5.5 xhigh Claude Code 79.0 $0.11 4,000
Claude Sonnet 5.5 high Claude Code 78.0 $0.06 2,000
Claude Haiku 5.5 max Claude Code 77.0 $0.09 45,000
Claude Opus 4.8 high Claude Code 76.0 $0.26 4,000
GPT-5.6 Sol high Codex 76.0 $0.21 4,000
DeepSeek V4.1 Flash max Codex 76.0 $0.08 41,000
DeepSeek V4.1 Flash max Claude Code 76.0 $0.20 48,000
Claude Haiku 5.5 xhigh Claude Code 76.0 $0.02 12,000
Claude Haiku 5.5 high Claude Code 76.0 $0.009 6,000
Kimi K3 Reasoning Kimi Code 75.0 $0.32 9,000
Claude Opus 4.7 high Claude Code 74.0 $0.34 4,000
GLM-5.2 max Claude Code 74.0 $0.34 28,000
Claude Sonnet 5 high Claude Code 74.0 $0.20 6,000
GLM-5.3 max Claude Code 74.0 $0.48 15,000
Claude Opus 5.5 low Claude Code 74.0 $0.08 1,000
GPT-5.6 Luna xhigh Codex 73.0 $0.02 7,000
Kimi K3 Reasoning Terminus-2 73.0 $0.23 6,000
GPT-6 Luna xhigh Codex 73.0 $0.007 6,000
DeepSeek-V4-Flash high Terminus-2 72.0 $0.07 31,000
DeepSeek-V4-Pro max Terminus-2 72.0 $0.35 24,000
Kimi K2.7 Code Reasoning Kimi Code 72.0 $0.18 11,000
Kimi K2.7 Code Reasoning Terminus-2 72.0 $0.11 12,000
GLM-5.2 max Terminus-2 72.0 $0.32 30,000
GPT-5.6 Luna high Codex 72.0 $0.02 5,000
GLM-5.3 max Terminus-2 72.0 $0.38 19,000
GLM-5.3-Flash max Claude Code 72.0 $0.09 29,000
Gemini 3.8 Flash low Terminus-2 72.0 $0.11 8,000
Claude Haiku 5.5 medium Claude Code 72.0 $0.006 4,000
GLM-5.2 high Terminus-2 71.0 $0.22 22,000
MiMo-V2.6 Flash Reasoning MiMo Code 71.0 $0.05 32,000
GPT-6 Luna high Codex 71.0 $0.004 3,000
Claude Sonnet 5.5 medium Claude Code 71.0 $0.05 1,000
Kimi K2.6 Reasoning Kimi Code 70.0 $0.27 18,000
Kimi K2.6 Reasoning Terminus-2 70.0 $0.09 12,000
DeepSeek-V4-Pro high Terminus-2 70.0 $0.30 20,000
Claude Haiku 5.5 low Claude Code 70.0 $0.005 3,000
DeepSeek-V4-Flash max Terminus-2 69.0 $0.06 29,000
GPT-5.6 Terra high Codex 69.0 $0.10 3,000
Qwen 3.8 27B Reasoning Qwen Code 69.0 $0.22 28,000
Claude Sonnet 5.5 low Claude Code 69.0 $0.04 1,000
MiniMax-M3 Reasoning MiniMax Code 68.0 $0.15 30,000
Nemotron 3 Ultra high Terminus-2 68.0 $0.08 9,000
GLM-5.3 low Terminus-2 68.0 $0.27 10,000
Qwen 3.8 27B xhigh Terminus-2 67.0 $0.12 22,000
DeepSeek-V4-Flash high Codex 66.0 $0.09 19,000
GPT-5.6 Luna medium Codex 66.0 $0.009 3,000
Muse Glimmer high Terminus-2 64.0 $0.08 11,000
GLM-5.2 Non-reasoning Terminus-2 63.0 $0.10 6,000
GLM-5.3-Flash low Terminus-2 63.0 $0.05 15,000
GLM-5.3-Flash max Terminus-2 63.0 $0.09 30,000
Mistral Large 4 high Terminus-2 63.0 $0.10 14,000
GPT-5.6 Luna low Codex 61.0 $0.006 2,000
Qwen 3.6 35B-A3B Reasoning Terminus-2 60.0 $0.07 16,000
DeepSeek-V4-Flash high Claude Code 59.0 $0.08 19,000
Claude Haiku 4.5 Reasoning Claude Code 58.0 $0.09 7,000
Qwen 3.6 35B-A3B Reasoning Qwen Code 57.0 18,000
GPT-6 Luna medium Codex 57.0 $0.003 1,000
Gemini 3.1 Flash-Lite high Terminus-2 56.0 $0.03 16,000
Gemma 4 26B A4B Reasoning Terminus-2 56.0 $0.02 21,000
Gemma 4 31B IT Reasoning Terminus-2 55.0 $0.02 8,000
MiniMax-M3.1-Flash max Terminus-2 53.0 $0.02 19,000
Step 5 Preview Reasoning Terminus-2 52.0 $0.13 24,000
Nemotron 3.5 Super VL Reasoning Terminus-2 52.0 $0.01 7,000
GPT-6 Luna low Codex 51.0 $0.003 1,000
Gemini 3.1 Flash-Lite low Terminus-2 48.0 $0.02 8,000
gpt-oss-120b high Terminus-2 46.0 $0.03 15,000
gpt-oss-120b high Codex 43.0 $0.26 27,000
glm-4.7-flash Reasoning Terminus-2 43.0 $0.04 14,000
glm-4.7-flash Reasoning Claude Code 41.0 $0.27 15,000
Laguna XS 2.1 Reasoning Terminus-2 37.0 $0.05 24,000
gpt-oss-20b high Terminus-2 33.0 $0.02 40,000
Mistral Large 3 Non-reasoning Terminus-2 33.0 $0.03 4,000
Gemini 3.1 Flash-Lite minimal Terminus-2 32.0 $0.007 2,000
gpt-oss-120b low Terminus-2 31.0 $0.005 2,000
Nemotron 3 Nano Omni 30B-A3B Reasoning Terminus-2 21.0 21,000
Nemotron 3.5 Lightning 30B-A3B Reasoning Terminus-2 13.0 $0.01 7,000
gpt-oss-20b high Codex 9.0 $0.09 23,000
Llama 3.3 70B Non-reasoning Terminus-2 8.0 $0.02 2,000
Nemotron 3.5 Bolt E4B Reasoning Terminus-2 4.0 14,000

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard 1 leader

Source: Official Spatial Agent Index v2 leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
0 20 40 60 80 100 2025 2026 2027 0 20 40 60 80 100 2025 2026 2027

Source: Official Spatial Agent Index v2 leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

More in Tool use

All 265 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Neon 56.3%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (48 entries).
  • Pass Models scored 47 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from October 2026, for Nemotron 3.5 Bolt E4B.
  • Pass Frontier models 4 of the top 10 have a score, led by Claude Opus 5.5 (#5).
  • Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on axisspatial.com: 91 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check