OSWorld Verified

🏆 Leaderboard

As of August 31, 2026, Qwen3.8 Max is #1 for OSWorld Verified at 86.1%. Ranked by OSWorld-Verified: cleaned desktop computer-use tasks with an execution check. 37 models in this index have a published OSWorld Verified score. Methodology: OSWorld (https://os-world.github.io/). OSWorld Verified leaderboard with live API prices. OSWorld Verified — validated tasks testing AI agents on real computer desktop environments. Official methodology: OSWorld (https://os-world.github.io/).

Updated August 31, 2026345 models36 providers

OSWorld Verified vs price

37 models. Left is cheaper. Up is a higher score. The line is the best score you can buy at each price — a dot under it is a worse deal than something on the line.

Best score at each price#1 on this board


Qwen3.8 MaxOSS Qwen · Open Source	86.1%	$8.00
Claude Mythos 5 Anthropic · Proprietary	85%	$60.00
Claude Fable 5 Anthropic · Proprietary	85%	$60.00
Qwen3.8-27BOSS Qwen · Open Source	84.3%	$3.65
Claude Opus 4.8 Anthropic · Proprietary	83.4%	$30.00
Claude Opus 5 Anthropic · Proprietary	83.4%	$30.00
Gemini 3.6 Flash Google · Proprietary	83%	$4.50
Claude Sonnet 5 Anthropic · Proprietary	81.2%	$12.00
Muse Spark 1.1 Meta · Proprietary	80.8%	$5.50
Claude Mythos Preview Anthropic · Proprietary	79.6%	$60.00
GPT-5.5 OpenAI · Proprietary	78.7%	$35.00
Gemini 3.5 Flash Google · Proprietary	78.4%	$10.50
Gemini 3.1 Pro Google · Proprietary	78.3%	$17.50
Claude Opus 4.7 Anthropic · Proprietary	78%	$30.00
GPT-5.4 OpenAI · Proprietary	75%	$17.50
Gemini 3.5 Flash-Lite Google · Proprietary	74%	$2.80
Qwen3.7-Plus Qwen · Proprietary	73.3%	$1.60
Kimi K2.6OSS Moonshot AI · Open Source	73.1%	$4.93
Claude Sonnet 4.6 Anthropic · Proprietary	72.1%	$18.00
GPT-5.4 mini OpenAI · Proprietary	72.1%	$5.25
MiniMax M3OSS MiniMax · Open Source	70.1%	$1.50
Muse Glimmer-30BOSS Meta · Open Source	65.9%	$1.85
GPT-5.3 Codex OpenAI · Proprietary	64.7%	$15.75
Kimi K2.5OSS Moonshot AI · Open Source	63.3%	$3.68
Claude Sonnet 4.5 Anthropic · Proprietary	62.9%	$18.00
Qwen3.6 Plus Qwen · Proprietary	62.5%	$3.50
Qwen3.5-122B-A10BOSS Qwen · Open Source	58%	$3.60
Qwen3.5-27BOSS Qwen · Open Source	56.2%	$2.70
Qwen3.5-35B-A3BOSS Qwen · Open Source	54.5%	$2.25
GPT-5 OpenAI · Proprietary	53.6%	$11.25
Gemini 3 Pro Google · Proprietary	47.5%	$14.00
Gemini 2.5 Pro Google · Proprietary	45.8%	$11.25
GPT-5.4 nano OpenAI · Proprietary	39%	$1.45
Claude 3.7 Sonnet Anthropic · Proprietary	35.8%	$18.00
o3 OpenAI · Proprietary	23%	$10.00
Qwen2.5 VL 72B InstructOSS Qwen · Open Source · via OpenRouter	5%	$1.80
Qwen2.5 VL 32B InstructOSS Qwen · Open Source	3.9%	$0.90
ChatGPT-4o Latest OpenAI · Proprietary	—	$12.50
Claude 3 Haiku Anthropic · Proprietary	—	$1.50
Claude 3 Opus Anthropic · Proprietary	—	$90.00
Claude 3 Sonnet Anthropic · Proprietary	—	$18.00
Claude 3.5 Haiku Anthropic · Proprietary	—	$4.80
Claude 3.5 Sonnet Anthropic · Proprietary	—	$18.00
Claude 3.5 Sonnet Anthropic · Proprietary	—	$18.00
Claude Haiku 4.5 Anthropic · Proprietary	—	$6.00
Claude Opus 4 Anthropic · Proprietary	—	$90.00
Claude Opus 4.1 Anthropic · Proprietary	—	$90.00
Claude Opus 4.5 Anthropic · Proprietary	—	$30.00
Claude Opus 4.6 Anthropic · Proprietary	—	$30.00
Claude Sonnet 4 Anthropic · Proprietary	—	$18.00

Showing 1–50 of 345 models

Next step

You found the model. Now ship the product.

Auth, billing, and the API layer are already decided. Fork 8 finished AI apps, or have us build the first version with you.

Build with the APIs Have us build it with you

The index

All Large Language Models

345 models across 36 providers. Search or jump to a lab — every model page stays linked here.

Cursor

2 models

Composer 2 Composer 2 Fast

xAI

17 models

Grok 4.6 Grok 4.5 Grok 4.3 Grok-4.1 Grok-4.1 Fast Non-Reasoning Grok-4.1 Fast Reasoning Grok-4.1 Thinking Grok 4 Fast Grok Code Fast 1 Grok-4 Fast Non-Reasoning Grok-4 Fast Reasoning Grok-4 Grok-3 Grok-3 Mini Grok-2 Grok-1.5V Grok-1.5

MiniMax

7 models

MiniMax M3 MiniMax M2.7 MiniMax M2.5 MiniMax M2.1 MiniMax M2 MiniMax M1 40K MiniMax M1 80K

Qwen

55 models

Qwen3.8 Flash Qwen3.8-Flash-Next Qwen3.8-27B Qwen3.8-2.4T-A95B Qwen3.8 Max Qwen3.7 Flash Qwen3.7-Plus Qwen3.7 Max Qwen3.6-27B Qwen3.6-35B-A3B Qwen3.6 Plus Qwen3.5-0.8B Qwen3.5-2B Qwen3.5-4B Qwen3.5-9B Qwen3.5-122B-A10B Qwen3.5-27B Qwen3.5-35B-A3B Qwen3.5-397B-A17B Qwen3 Max Qwen3 VL 235B A22B Instruct Qwen3 VL 235B A22B Thinking Qwen3 VL 30B A3B Instruct Qwen3 VL 30B A3B Thinking Qwen3 VL 32B Instruct Qwen3 VL 32B Thinking Qwen3 VL 4B Instruct Qwen3 VL 4B Thinking Qwen3 VL 8B Instruct Qwen3 VL 8B Thinking Qwen3-Next-80B-A3B-Instruct Qwen3-Next-80B-A3B-Thinking Qwen3-235B-A22B-Thinking-2507 Qwen3-235B-A22B-Instruct-2507 Qwen3 235B A22B Qwen3 30B A3B Qwen3 32B Qwen2.5-Omni-7B QwQ-32B Qwen2.5 VL 32B Instruct Qwen3-Coder 480B A35B Instruct Qwen2.5 VL 72B Instruct Qwen2.5 VL 7B Instruct Qwen3-Coder QvQ-72B-Preview QwQ-32B-Preview Qwen2.5 14B Instruct Qwen2.5 32B Instruct Qwen2.5 72B Instruct Qwen2.5 7B Instruct Qwen2.5-Coder 32B Instruct Qwen2.5-Coder 7B Instruct Qwen2-VL-72B-Instruct Qwen2 72B Instruct Qwen2 7B Instruct

AI21 Labs

2 models

Jamba 1.5 Large Jamba 1.5 Mini

Amazon

7 models

Nova 2 Lite Nova 2 Omni Nova 2 Pro Nova 2 Sonic Nova Lite Nova Micro Nova Pro

Baidu

2 models

ERNIE 5.0 ERNIE 4.5

ByteDance

4 models

Seed 2.1 Pro Seed 2.1 Turbo Seed 2.0 Lite Seed 2.0 Pro

Cohere

4 models

North Micro Vision Instruct North Mini Code 1.0 Command A+Command R+

IBM

2 models

IBM Granite 4.0 Tiny Preview Granite 3.3 8B Instruct

Inception

1 models

Mercury 2

inclusionAI

1 models

Ling-3.0-flash

LG AI Research

1 models

K-EXAONE-236B-A23B

Liquid AI

2 models

LFM2.5-VL-3B LFM2.5-2.6B

Meituan

4 models

LongCat-Flash-Lite LongCat-Flash-Thinking-2601 LongCat-Flash-Thinking LongCat-Flash-Chat

Microsoft

9 models

MAI-Code-1.1-Flash MAI-Code-1-Flash MAI-Thinking-1 Phi 4 Mini Phi-4-multimodal-instruct Phi 4 Phi-3.5-mini-instruct Phi-3.5-MoE-instruct Phi-3.5-vision-instruct

Moonshot AI

11 models

Kimi K3 Kimi K2.7 Code Kimi K2.7 Code HighSpeed Kimi K2.6 Kimi K2.5 Kimi K2 0905 Kimi K2-Instruct-0905 Kimi K2-Thinking-0905 Kimi K2 Base Kimi K2 Instruct Kimi-k1.5

Nous Research

1 models

Hermes 3 70B

NVIDIA

9 models

Nemotron 3.5 Lightning (30B A3B)Nemotron 3 Ultra (550B A55B)Nemotron 3 Super (120B A12B)Nemotron 3 Nano (30B A3B)Nemotron Nano 9B v2 Llama 3.1 Nemotron Ultra 253B v1 Llama 3.1 Nemotron Nano 8B V1 Llama-3.3 Nemotron Super 49B v1 Llama 3.1 Nemotron 70B Instruct

OpenBMB

1 models

MiniCPM-SALA

Poolside

2 models

Laguna S 2.1 Laguna XS 2.1

Sakana AI

1 models

Sakana Namazu

Sarvam AI

2 models

Sarvam-105B Sarvam-30B

StepFun

2 models

Step-3.5-Flash Step3-VL-10B

Tencent

1 models

Hy3

Thinking Machines

1 models

Inkling-Small

Unisound

1 models

Upstage

1 models

Solar Pro 4

Xiaomi

5 models

MiMo-V2.5-Pro MiMo-V2.5 MiMo-V2-Omni MiMo-V2-Pro MiMo-V2-Flash

Which model leads OSWorld Verified right now?

As of August 31, 2026, Qwen3.8 Max by Qwen is #1 for OSWorld Verified at 86.1%. Ranked by OSWorld-Verified: cleaned desktop computer-use tasks with an execution check. This board also tracks OSWorld Verified. Next on the same board: Claude Fable 5 and Claude Mythos 5. This osworld verified leaderboard ranks models by OSWorld Verified. Scores come from public evals. Prices are the live API rates in the table above.

Sources: OSWorld (https://os-world.github.io/); OpenAI API pricing (https://developers.openai.com/api/docs/pricing); Anthropic Claude API pricing (https://platform.claude.com/docs/en/about-claude/pricing); Gemini API pricing (https://ai.google.dev/gemini-api/docs/pricing)

Top 8 for osworld verified. Ranked by OSWorld-Verified: cleaned desktop computer-use tasks with an execution check. Input and output are dollars per million tokens.
Rank	Model	OSWorld Verified	Input /M	Output /M
1	Qwen3.8 Max	86.1%	$2.00	$6.00
2	Claude Fable 5	85%	$10.00	$50.00
3	Claude Mythos 5	85%	$10.00	$50.00
4	Qwen3.8-27B	84.3%	$0.45	$3.20
5	Claude Opus 4.8	83.4%	$5.00	$25.00
6	Claude Opus 5	83.4%	$5.00	$25.00
7	Gemini 3.6 Flash	83%	$0.75	$3.75
8	Claude Sonnet 5	81.2%	$2.00	$10.00

Related LLM leaderboards

LLM Leaderboard Coding Math Agentic Reasoning Writing Science Finance Long Context Vision Open Source

Individual benchmark pages

Rank one eval at a time. All LLM benchmarks.

HLE SWE Bench Humanity's Last Exam (No Tools)Humanity's Last Exam (With Tools)Terminal-Bench 2.0 Terminal-Bench Terminal-Bench 2.1 Terminal-Bench 2 Terminal-Bench 3.0 ARC-AGI-1 Verified

OSWorld Verified FAQ

Who ranks #1 on the OSWorld Verified leaderboard?

As of August 31, 2026, Qwen3.8 Max by Qwen ranks #1 on OSWorld Verified at 86.1%. API pricing is $2.00/M input and $6.00/M output.

What are the top models on OSWorld Verified?

The current OSWorld Verified ranking as of August 31, 2026 is 1. Qwen3.8 Max at 86.1%; 2. Claude Fable 5 at 85%; 3. Claude Mythos 5 at 85%.

Which osworld verified model is the cheapest?

Qwen2.5 VL 32B Instruct is the cheapest scored model on this osworld verified leaderboard at $0.40/M input and $0.50/M output ($0.90 blended). Qwen3.8 Max still leads OSWorld Verified at 86.1%.

Should I always pick the #1 OSWorld Verified model?

Not automatically. Qwen3.8 Max leads OSWorld Verified, but a cheaper scored model can be the better production choice if the quality gap is small. Use the table to weigh OSWorld Verified against input/output price, context window, and related evals.

How often is the OSWorld Verified leaderboard updated?

Scores and API prices on this page are refreshed from published evals and provider rates. The snapshot is labeled August 31, 2026. Treat it as a current index, not a one-off blog post.

What is OSWorld Verified?

OSWorld Verified — validated tasks testing AI agents on real computer desktop environments. This page ranks models that have published a OSWorld Verified score, with live API token prices on the same row. Official methodology: OSWorld (https://os-world.github.io/).

Where is the OSWorld Verified leaderboard?

This page is the OSWorld Verified leaderboard. Models are sorted by OSWorld Verified, with input and output token prices on the same row so you can weigh score against cost. Official boards often omit price; that comparison is the point of this index.

How is this OSWorld Verified ranking different from the official board?

The official OSWorld page owns the methodology. This page keeps the published OSWorld Verified score next to live API $/M so you can pick a production SKU, not only a trophy number. Source: https://os-world.github.io/

OSWorld Verified vs price

You found the model. Now ship the product.

All Large Language Models

OpenAI

Anthropic

Google

Cursor

xAI

DeepSeek

MiniMax

Qwen

AI21 Labs

Amazon

Baidu

ByteDance

Cohere

IBM

Inception

inclusionAI

LG AI Research

Liquid AI

Meituan

Meta

Microsoft

Mistral

Moonshot AI

Nous Research

NVIDIA

OpenBMB

Poolside

Sakana AI

Sarvam AI

StepFun

Tencent

Thinking Machines

Unisound

Upstage

Xiaomi

Z AI

Which model leads OSWorld Verified right now?

Related LLM leaderboards

Individual benchmark pages

OSWorld Verified FAQ

Who ranks #1 on the OSWorld Verified leaderboard?

What are the top models on OSWorld Verified?

Which osworld verified model is the cheapest?

Should I always pick the #1 OSWorld Verified model?

How often is the OSWorld Verified leaderboard updated?

What is OSWorld Verified?

Where is the OSWorld Verified leaderboard?

How is this OSWorld Verified ranking different from the official board?