OpenHands Index

Name: OpenHands Index results
Creator: OpenHands Index
License: https://www.apache.org/licenses/LICENSE-2.0

The average of five coding-agent tasks, weighted equally: resolving GitHub issues, front-end work, building a project from scratch, writing tests and gathering information.

Data older than 90 days: last updated 30 Jun 2026

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

Average: DeepSeek-V4-Pro (0423)

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

1 Claude Fable 5Anthropic 81.0%
2 Claude Opus 4.8Anthropic 71.9%
3 Claude Opus 4.7Anthropic 69.7%
4 Claude Opus 4.6Anthropic 66.7%
5 GPT-5.5OpenAI 65.9%
6 GPT-5.4OpenAI 64.3%
7 Gemini 3.5 FlashGoogle 62.6%
8 Claude Opus 4.5Anthropic 60.6%
8 Gemini 3.1 Pro PreviewGoogle 60.6%
10 GPT-5.2OpenAI 58.8%
11 GPT-5.2-CodexOpenAI 58.3%
12 GLM-5.1Z.ai 58.2%
13 MiniMax-M3MiniMax 57.2%
14 Kimi K2.6Moonshot AI 57.1%
15 Claude Sonnet 4.5Anthropic 53.0%
⋯
29 DeepSeek-V4-Pro (0423)DeepSeek 40.7%

Full results

OpenHands Index: average, % resolved, higher is better
#	Model	Average % resolved, higher is better	Price $ per million tokens, in / out
1	Claude Fable 5Anthropic	81.0%	$10 / $50
2	Claude Opus 4.8Anthropic	71.9%	$5 / $25
3	Claude Opus 4.7Anthropic	69.7%	$5 / $25
4	Claude Opus 4.6Anthropic	66.7%	$5 / $25
5	GPT-5.5OpenAI	65.9%	$5 / $30
6	GPT-5.4OpenAI	64.3%	$2.50 / $15
7	Gemini 3.5 FlashGoogle	62.6%	$1.50 / $9
8	Claude Opus 4.5Anthropic	60.6%	$5 / $25
8	Gemini 3.1 Pro PreviewGoogle	60.6%	$2 / $12
10	GPT-5.2OpenAI	58.8%	$1.75 / $14
11	GPT-5.2-CodexOpenAI	58.3%	$1.75 / $14
12	GLM-5.1Z.ai	58.2%	$1.38 / $4.40
13	MiniMax-M3MiniMax	57.2%	$0.30 / $1.20
14	Kimi K2.6Moonshot AI	57.1%	$0.95 / $4
15	Claude Sonnet 4.5Anthropic	53.0%	$3 / $15
16	Qwen3.6-PlusAlibaba	52.9%	$0.33 / $1.95
17	GLM-5Z.ai	49.4%	$0.95 / $2.55
18	Kimi K2.5Moonshot AI	49.2%	$0.57 / $2.85
19	Gemini 3 Pro PreviewGoogle	49.0%	–
19	Gemini 3 Flash PreviewGoogle	49.0%	$0.50 / $3
21	DeepSeek-V3.2 (reasoning on)DeepSeek	45.7%	$0.30 / $0.96
22	MiniMax-M2.5MiniMax	45.2%	$0.30 / $1.20
23	Claude Sonnet 4.6Anthropic	44.5%	$3 / $15
24	Qwen3-Coder-NextAlibaba	43.8%	$0.18 / $0.90
25	MiniMax-M2.7MiniMax	43.4%	$0.30 / $1.20
26	GLM-4.7Z.ai	42.3%	$0.54 / $1.98
27	MiniMax-M2.1MiniMax	41.2%	$0.30 / $1.20
28	Kimi K2 ThinkingMoonshot AI	41.0%	$0.60 / $2.50
29	DeepSeek-V4-Pro (0423)DeepSeek	40.7%	$1.42 / $2.83
30	Qwen3.5-FlashAlibaba	38.1%	$0.065 / $0.26
31	Nemotron 3 Super 120B A12BNVIDIA	36.2%	$0.085 / $0.40
32	Trinity Large ThinkingArcee AI	32.1%	$0.25 / $0.80
33	Qwen3-Coder-480B-A35BAlibaba	30.9%	$0.35 / $1.50
34	Nemotron 3 Nano 30B A3BNVIDIA	15.5%	$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

Average, 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.

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.

AnthropicOpenAIGoogleZ.aiMiniMaxMoonshot AI

What it measures

The average of five coding-agent tasks, weighted equally: resolving GitHub issues, front-end work, building a project from scratch, writing tests and gathering information.

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.