Language modelOpen weights

Ling-3.0-flash-VL

InclusionAI

Released: -
Data date: October 3, 2026

Ling-3.0-flash-VL adds native image and video input to Ling 3.0 Flash. VideoRoPE represents spatial positions and temporal order, supporting questions about events in longer videos. InclusionAI adds a visual encoder to the language backbone for this purpose.

The MoE model activates 5.5 of its 124 billion parameters per token and handles up to 256K context tokens. Its local weights use MIT.

Specifications and access

Specification	Value and source
Model ID	inclusionAI/Ling-3.0-flash-VLSource
Model class	Multimodal language model with a vision encoderSource
Parameters	124 billion total, 5.5 billion activeSource
Context window	Up to 256K tokens according to the model cardSource
Architecture	ViT vision encoder, MLP projector, VideoRoPE, and sparse MoE backboneSource
Input	Text, images, and videoSource
License	MITSource

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

Artificial Analysis Intelligence Index v4.1.1 (Show measurement, test conditions, and source)

Source value: 42
Score: 42
Metric: Artificial Analysis Intelligence Index v4.1.1 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

CountBench (Show measurement, test conditions, and source)

Source value: 97.33
Score: 97.33
Metric: CountBench reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

WorldVQA (Show measurement, test conditions, and source)

Source value: 45.67
Score: 45.67
Metric: WorldVQA reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

MMMU-Pro (Show measurement, test conditions, and source)

Source value: 79
Score: 79
Metric: MMMU-Pro reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

MathVision (Show measurement, test conditions, and source)

Source value: 84.87
Score: 84.87
Metric: MathVision reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

Humanity's Last Exam-MM (Show measurement, test conditions, and source)

Source value: 19.88
Score: 19.88
Metric: Humanity's Last Exam-MM reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

OmniDocBench 1.5 (Show measurement, test conditions, and source)

Source value: 91.35
Score: 91.35
Metric: OmniDocBench 1.5 reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

CharXiv_RQ (Show measurement, test conditions, and source)

Source value: 81.3
Score: 81.3
Metric: CharXiv_RQ reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

MMSearch (Show measurement, test conditions, and source)

Source value: 79
Score: 79
Metric: MMSearch reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

ClawEval-MM (Show measurement, test conditions, and source)

Source value: 59.9
Score: 59.9
Metric: ClawEval-MM reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

WebVoyager (Show measurement, test conditions, and source)

Source value: 90.83
Score: 90.83
Metric: WebVoyager reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

Vision2Web (Show measurement, test conditions, and source)

Source value: 57.69
Score: 57.69
Metric: Vision2Web reported score
Unit: No unit provided
Category: source-specific
Direction: Higher is better
Source type: vendor-reported
Evaluator: Provider model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Official image-only table, visually inspected. Default thinking T=0.6/top-p=.95/top-k=20; TerminalBench note specifies AA Terminus 2, 2h, three-run mean, T=1.0. AntBench-Medical is in-house and not released.
Context: Source-specific observation; it is not a shared comparison cohort. The source table does not state a unit or scale for this numeric score.

Source Provider model card

Sources and data date

Every statement links to its underlying documentation or leaderboard.

Type	Evidence and data date
Research status	Research date October 4, 2026. 1 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Unresolved	antbench-medical-score-components Original wording The source presents two AntBench-Medical numbers without assigning each number to a named metric, so neither is imported as a benchmark result.
Additional source	Ling-3.0-flash-VL model card (retrieved October 3, 2026; October 4, 2026) · Ling-3.0-flash-VL model card · Editorial description reviewed October 4, 2026
Additional source	Ling-3.0-flash base model card (retrieved October 3, 2026) · Ling-3.0-flash base model card · Editorial description reviewed October 4, 2026
Additional source	Ling-3.0-flash-Fin model card (retrieved October 3, 2026) · Ling-3.0-flash-Fin model card · Editorial description reviewed October 4, 2026