Benchmarks / OpenHands Index

Reported by OpenHands Index

OpenHands Index

Resolving real GitHub issues so that the repository's own tests pass.

Data older than 90 days: last updated 30 Jun 2026

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

Issue resolution (SWE-Bench): GLM-5.1

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

  1. 1 Claude Fable 5Anthropic 95.8%
  2. 2 Claude Opus 4.8Anthropic 83.8%
  3. 3 Claude Opus 4.7Anthropic 81.6%
  4. 4 Gemini 3.5 FlashGoogle 78.6%
  5. 5 GPT-5.5OpenAI 78.2%
  6. 6 Claude Opus 4.6Anthropic 76.8%
  7. 6 Gemini 3.1 Pro PreviewGoogle 76.8%
  8. 8 Claude Opus 4.5Anthropic 76.6%
  9. 9 MiniMax-M3MiniMax 76.4%
  10. 10 MiniMax-M2.7MiniMax 75.6%
  11. 10 GPT-5.4OpenAI 75.6%
  12. 12 GLM-5.1Z.ai 75.0%
  13. 13 Gemini 3 Flash PreviewGoogle 74.6%
  14. 13 Kimi K2.6Moonshot AI 74.6%
  15. 13 GPT-5.2OpenAI 74.6%

Full results

OpenHands Index: issue resolution (SWE-Bench), % resolved, higher is better
#ModelIssue resolution (SWE-Bench)
% resolved, higher is better		Price
$ per million tokens, in / out
1 Claude Fable 5Anthropic
95.8%
$10 / $50
2 Claude Opus 4.8Anthropic
83.8%
$5 / $25
3 Claude Opus 4.7Anthropic
81.6%
$5 / $25
4 Gemini 3.5 FlashGoogle
78.6%
$1.50 / $9
5 GPT-5.5OpenAI
78.2%
$5 / $30
6 Claude Opus 4.6Anthropic
76.8%
$5 / $25
6 Gemini 3.1 Pro PreviewGoogle
76.8%
$2 / $12
8 Claude Opus 4.5Anthropic
76.6%
$5 / $25
9 MiniMax-M3MiniMax
76.4%
$0.30 / $1.20
10 MiniMax-M2.7MiniMax
75.6%
$0.30 / $1.20
10 GPT-5.4OpenAI
75.6%
$2.50 / $15
12 GLM-5.1Z.ai
75.0%
$1.38 / $4.40
13 Gemini 3 Flash PreviewGoogle
74.6%
$0.50 / $3
13 Kimi K2.6Moonshot AI
74.6%
$0.95 / $4
13 GPT-5.2OpenAI
74.6%
$1.75 / $14
16 Claude Sonnet 4.6Anthropic
74.4%
$3 / $15
17 Qwen3.6-PlusAlibaba
74.2%
$0.33 / $1.95
17 Claude Sonnet 4.5Anthropic
74.2%
$3 / $15
19 GPT-5.2-CodexOpenAI
73.8%
$1.75 / $14
20 GLM-4.7Z.ai
73.4%
$0.54 / $1.98
20 GLM-5Z.ai
73.4%
$0.95 / $2.55
22 DeepSeek-V4-Pro (0423)DeepSeek
73.2%
$1.42 / $2.83
23 MiniMax-M2.5MiniMax
72.6%
$0.30 / $1.20
24 DeepSeek-V3.2 (reasoning on)DeepSeek
71.6%
$0.30 / $0.96
25 Gemini 3 Pro PreviewGoogle
70.6%
–
26 Kimi K2 ThinkingMoonshot AI
69.2%
$0.60 / $2.50
27 MiniMax-M2.1MiniMax
68.8%
$0.30 / $1.20
27 Kimi K2.5Moonshot AI
68.8%
$0.57 / $2.85
29 Qwen3-Coder-NextAlibaba
66.6%
$0.18 / $0.90
30 Qwen3-Coder-480B-A35BAlibaba
62.4%
$0.35 / $1.50
31 Qwen3.5-FlashAlibaba
62.0%
$0.065 / $0.26
31 Nemotron 3 Super 120B A12BNVIDIA
62.0%
$0.085 / $0.40
33 Trinity Large ThinkingArcee AI
56.8%
$0.25 / $0.80
34 Nemotron 3 Nano 30B A3BNVIDIA
34.2%
$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

Issue resolution (SWE-Bench), 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.

20%40%60%80%100% Oct 20252026AprJulOct Qwen3-Coder-480B-A35BClaude Sonnet 4.5Claude Opus 4.5Claude Opus 4.6Claude Opus 4.7Claude Opus 4.8Claude Fable 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.

AnthropicGoogleOpenAIMiniMaxZ.aiMoonshot AI

60%70%80%90%100% Oct 20252026AprJulOct AnthropicGoogleOpenAIMiniMaxZ.aiMoonshot AI

What it measures

Resolving real GitHub issues so that the repository's own tests pass.

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.