OSWorld-G

Fara1.5-27B 71.3% leads since June 2026

OpenAI Computer-Using Agent (CUA) 40.6% leads since May 2025

Fara1.5-27B 71.3% leads since June 2026

Fading

Labs’ own numbers All independent runs Include claimed scores

Models map 564 desktop instructions to the right screen element or point, covering text matching, element recognition, layout understanding and fine manipulation.

Released with the Jedi grounding dataset. NeurIPS 2025 spotlight. Scored by click accuracy.

Made by: XLANG Lab, University of Hong Kong (OSWorld team)
Released: May 2025
Kind: Finding things on screen
Level: Everyday
Scored: Checked automatically
Answers: Short answer
Tasks: All public

Last in a launch post: January 2026, dropped before it saturated
Last model added: June 2026
Scores refreshed: November 12th, 2026

Models scored in lab cards

Launch posts list it, from 4 labs

System cards report it, from 5 labs

Fara1.5-27B leads with 71.3% among models released by May 2026

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

OpenAI Computer-Using Agent (CUA) leads with 40.6% across all independent runs

Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.

Fara1.5-27B leads with 71.3%, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

Bars Table

The newest model with a score here came out in May 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

Source: Labs’ model cards

1 OpenAI Computer-Using Agent (CUA) 40.6%
2 Qwen2.5-VL-7B 31.4%
3 Qwen2.5-VL-3B 27.3%

Source: Official OSWorld-G leaderboard

1 Fara1.5-27B 71.3%
2 Qwen3-VL-235B-A22B 68.3%
3 STEP3-VL-10B 59%
4 InternVL3.5-241B-A28B 53.2%
5 Kimi-VL-A3B-Thinking-2506 52.5%
6 OpenAI Computer-Using Agent (CUA) 40.6%
7 Qwen2.5-VL-7B 31.4%
8 Qwen2.5-VL-3B 27.3%

Sources: Labs’ model cards and official OSWorld-G leaderboard

#	Model	Score	Date	Source
1	Fara1.5-27B	71.3%	Jun 2026	Lab model card
2	Qwen3-VL-235B-A22B	68.3%	Sep 2025	Lab model card
3	STEP3-VL-10B	59%	Jan 2026	Lab model card
4	InternVL3.5-241B-A28B	53.2%	Aug 2025	Lab model card
5	Kimi-VL-A3B-Thinking-2506	52.5%	Jun 2025	Lab model card

#	Model	Score	Date	Source
1	OpenAI Computer-Using Agent (CUA)	40.6%	Jan 2025	Official leaderboard
2	Qwen2.5-VL-7B	31.4%	Jan 2025	Official leaderboard
3	Qwen2.5-VL-3B	27.3%	Jan 2025	Official leaderboard

#	Model	Score	Date	Source
1	Fara1.5-27B	71.3%	Jun 2026	Lab model card
2	Qwen3-VL-235B-A22B	68.3%	Sep 2025	Lab model card
3	STEP3-VL-10B	59%	Jan 2026	Lab model card
4	InternVL3.5-241B-A28B	53.2%	Aug 2025	Lab model card
5	Kimi-VL-A3B-Thinking-2506	52.5%	Jun 2025	Lab model card
6	OpenAI Computer-Using Agent (CUA)	40.6%	Jan 2025	Official leaderboard
7	Qwen2.5-VL-7B	31.4%	Jan 2025	Official leaderboard
8	Qwen2.5-VL-3B	27.3%	Jan 2025	Official leaderboard

The benchmark's own leaderboard has 7 entries, 3 of them models we track. Switch the scores above to rank them together.

4 models have held number one in labs’ own numbers since its release in May 2025 Number one over time 5 models have held number one with claims included since its release in May 2025

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.

OpenAI Moonshot AI Shanghai AI Lab Qwen Microsoft

Own leaderboard Not tested since May 2025

OpenAI Computer-Using Agent (CUA) 40.6%

Lab claims Not tested since Jun 2026

Fara1.5-27B 71.3%

Sources: Official OSWorld-G leaderboard and labs’ model cards

Scores over time

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

On the benchmark’s own leaderboard Reported by the lab in its model card Best so far

Sources: Official OSWorld-G leaderboard and labs’ model cards

Labs headline it less often now, last in January 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Moonshot AI

Shanghai AI Lab

Qwen

StepFun

Date	Launch post	Score	As the post writes it
Jan 13, 2026	STEP3-VL-10B	59	OSWorld-G
Sep 23, 2025	Qwen3-VL-235B-A22B	68.3	OSWorldG
Aug 25, 2025	InternVL3.5-241B-A28B	53.2	OSWorld-G
Jun 21, 2025	Kimi-VL-A3B-Thinking-2506	52.5	OSWorld-G (Acc, full set with refusal)

5 system cards report it, from 5 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

Date	Document	Score	As the document writes it	Section
Jun 18, 2026	Fara1.5-27B technical report	71.3	OSWorld-G Refined grounding accuracy (Table 6)	Capabilities
Jan 14, 2026	STEP3-VL-10B technical report	59	OSWorld-G, SeRe (Table 1)	Capabilities
Nov 26, 2025	Qwen3-VL-235B-A22B technical report	68.3	OSWorldG (Table 2)	Capabilities
Aug 25, 2025	InternVL3.5-241B-A28B technical report	53.2	OSWorld-G (Table 2, Table 10)	Capabilities
Jun 23, 2025	Kimi-VL-A3B-Thinking-2506 technical report	52.5	OSWorld-G (Acc), full set with refusal samples (Table 5, section 4.3)	Capabilities