Benchmarks / OpenHands Index

Reported by OpenHands Index

OpenHands Index

Answering questions that need web browsing and tools.

Data older than 90 days: last updated 30 Jun 2026

Results dated
30 Jun 2026
Models
34
Unit
% resolved
Licence
Apache License 2.0

Information gathering (GAIA): Qwen3-Coder-Next

Top 15 of 34 results · % resolved, higher is better. Choose a model to highlight it.Clear highlight

  1. 1 GPT-5.5OpenAI 86.1%
  2. 2 Claude Fable 5Anthropic 84.2%
  3. 3 GPT-5.4OpenAI 82.4%
  4. 4 Gemini 3.1 Pro PreviewGoogle 81.8%
  5. 5 Claude Opus 4.7Anthropic 81.2%
  6. 5 Gemini 3.5 FlashGoogle 81.2%
  7. 7 Claude Opus 4.6Anthropic 80.0%
  8. 8 Claude Opus 4.8Anthropic 78.8%
  9. 9 Kimi K2.6Moonshot AI 74.5%
  10. 10 Claude Sonnet 4.5Anthropic 72.7%
  11. 11 Qwen3.6-PlusAlibaba 72.1%
  12. 12 GPT-5.2-CodexOpenAI 70.9%
  13. 13 Claude Opus 4.5Anthropic 69.1%
  14. 14 GLM-5.1Z.ai 67.3%
  15. 15 MiniMax-M3MiniMax 66.7%
  16. 21 Qwen3-Coder-NextAlibaba 50.9%

Full results

OpenHands Index: information gathering (GAIA), % resolved, higher is better
#ModelInformation gathering (GAIA)
% resolved, higher is better		Price
$ per million tokens, in / out
1 GPT-5.5OpenAI
86.1%
$5 / $30
2 Claude Fable 5Anthropic
84.2%
$10 / $50
3 GPT-5.4OpenAI
82.4%
$2.50 / $15
4 Gemini 3.1 Pro PreviewGoogle
81.8%
$2 / $12
5 Claude Opus 4.7Anthropic
81.2%
$5 / $25
5 Gemini 3.5 FlashGoogle
81.2%
$1.50 / $9
7 Claude Opus 4.6Anthropic
80.0%
$5 / $25
8 Claude Opus 4.8Anthropic
78.8%
$5 / $25
9 Kimi K2.6Moonshot AI
74.5%
$0.95 / $4
10 Claude Sonnet 4.5Anthropic
72.7%
$3 / $15
11 Qwen3.6-PlusAlibaba
72.1%
$0.33 / $1.95
12 GPT-5.2-CodexOpenAI
70.9%
$1.75 / $14
13 Claude Opus 4.5Anthropic
69.1%
$5 / $25
14 GLM-5.1Z.ai
67.3%
$1.38 / $4.40
15 MiniMax-M3MiniMax
66.7%
$0.30 / $1.20
16 GPT-5.2OpenAI
65.5%
$1.75 / $14
17 Kimi K2.5Moonshot AI
63.6%
$0.57 / $2.85
18 GLM-5Z.ai
60.0%
$0.95 / $2.55
19 Gemini 3 Flash PreviewGoogle
58.8%
$0.50 / $3
20 GLM-4.7Z.ai
53.9%
$0.54 / $1.98
21 Qwen3-Coder-NextAlibaba
50.9%
$0.18 / $0.90
22 DeepSeek-V3.2 (reasoning on)DeepSeek
50.3%
$0.30 / $0.96
23 Qwen3.5-FlashAlibaba
49.7%
$0.065 / $0.26
24 MiniMax-M2.5MiniMax
47.9%
$0.30 / $1.20
25 Gemini 3 Pro PreviewGoogle
44.2%
–
26 Kimi K2 ThinkingMoonshot AI
43.6%
$0.60 / $2.50
27 MiniMax-M2.1MiniMax
40.6%
$0.30 / $1.20
28 Nemotron 3 Super 120B A12BNVIDIA
40.0%
$0.085 / $0.40
29 Qwen3-Coder-480B-A35BAlibaba
33.9%
$0.35 / $1.50
30 Trinity Large ThinkingArcee AI
32.7%
$0.25 / $0.80
31 MiniMax-M2.7MiniMax
25.5%
$0.30 / $1.20
32 Claude Sonnet 4.6Anthropic
13.3%
$3 / $15
33 DeepSeek-V4-Pro (0423)DeepSeek
12.7%
$1.42 / $2.83
34 Nemotron 3 Nano 30B A3BNVIDIA
8.5%
$0.05 / $0.20

Swipe the table sideways for more columns.

Ranks follow the score as shown, so equal numbers share a rank. Results as published by OpenHands Index; we do not re-run them.

Progress over time

Information gathering (GAIA), plotted at each model's release date, higher is better. Each model counts once, at its best setting. Scores are today's, not launch day's; 1 model without a known release date is left off. Hover or tap a mark for its model.

Best score so far

Every model at its release date. The line moves each time a new release beats every earlier one, and each new best is named where there is room.

0%20%40%60%80%100% Oct 20252026AprJulOct Qwen3-Coder-480B-A35BClaude Sonnet 4.5Gemini 3.1 Pro PreviewGPT-5.4GPT-5.5

Best score by lab

Each lab's best so far, stepping whenever one of its releases beats its own earlier models. The 6 labs with the best scores today; 4 others not drawn.

OpenAIAnthropicGoogleMoonshot AIAlibabaZ.ai

20%40%60%80%100% Oct 20252026AprJulOct OpenAIAnthropicGoogleMoonshot AIAlibabaZ.ai

What it measures

Answering questions that need web browsing and tools.

What it does not measure

Not the model alone: the score is for the model inside the OpenHands agent, at the SDK version shown.

Source

OpenHands Index · Methodology

OpenHands Index by the OpenHands contributors, Apache License 2.0.