Benchmarks · /benchmarks/scale-vtb

VTB

Name: VTB
Creator: Scale Labs

Scale VTB leaderboard for image-grounded reasoning and multimodal understanding.

Source · Scale Labs
Version · scale-labs snapshot 2026-10-03
Scores · 25

Test details

Visible tradeoffsThis is a rubric-judged signal, so it is more structured than arena taste but still depends on the scoring rubric.

source

Scale Labs

metric

APR (%)

judge

Rubric

direction

higher better

group id

scale_vtb_current

domain

Vision understanding

What it measures vs what it misses

✓ Measures

Visual interpretation and reasoning over benchmark images and prompts.

✗ Misses

Image generation quality. Tool-use orchestration beyond the judged task.

Why this countsIt is useful when the model must read charts, UI, screenshots, or visual scenes rather than text alone.Same-test ruleThis percentile only compares models inside the exact benchmark/version group shown here. It is not a universal score.What it missesIt does not tell you whether the model can generate or edit images well.

Leaderboard · this benchmark version

#1 · muse-spark-1.1

SL · Oct 3, 2026

Source label: Muse Spark 1.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 100%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: Muse Spark 1.1. Collapse policy: highest reported score per canonical model.

44.8%

#2 · GPT-5.4

SL · Oct 3, 2026

Source label: gpt-5.4-2026-03-05 (reasoning effort = high)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 95.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5.4-2026-03-05 (reasoning effort = high). Collapse policy: highest reported score per canonical model.

29.2%

#3 · Gemini 3.1 Pro Preview

SL · Oct 3, 2026

Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 91.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-3.1-pro-preview. Collapse policy: highest reported score per canonical model.

29%

#4 · Claude Opus 4.6

SL · Oct 3, 2026

Source label: claude-opus-4-6-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 87.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-6-thinking. Collapse policy: highest reported score per canonical model.

27.5%

#5 · Gemini 3 Pro Preview

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 83.3%
Last updated: recent
Eligibility: preview_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-3-pro-preview. Collapse policy: highest reported score per canonical model.

26.9%

#6 · GPT-5

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5-2025-08-07-thinking. Collapse policy: highest reported score per canonical model.

18.7%

#7 · GPT-5.4 mini

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5-2025-08-07-thinking. Collapse policy: highest reported score per canonical model. Backfilled from GPT-5 via approved benchmark identity mapping map-gpt-5-4-mini-to-gpt-5.

18.7%

#8 · GPT-5.4 nano

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

18.7%

#9 · o3

SL · Oct 3, 2026

Source label: o3-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 66.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: o3-2025-04-16. Collapse policy: highest reported score per canonical model.

13.7%

#10 · Gemini 2.5 Pro

SL · Oct 3, 2026

Source label: gemini-2.5-pro-preview-06-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 62.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-2.5-pro-preview-06-05. Collapse policy: highest reported score per canonical model.

11.8%

#11 · o4 mini

SL · Oct 3, 2026

Source label: o4-mini-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 58.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: o4-mini-2025-04-16. Collapse policy: highest reported score per canonical model.

11.1%

#12 · claude-sonnet-4-5-20250929-thinking

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 54.2%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-5-20250929-thinking. Collapse policy: highest reported score per canonical model.

6.2%

#13 · Claude Sonnet 4.5

SL · Oct 3, 2026

Source label: claude-sonnet-4-5-20250929

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 50%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-5-20250929. Collapse policy: highest reported score per canonical model.

5.6%

#14 · GPT-4.1

SL · Oct 3, 2026

Source label: gpt-4.1-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 45.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-4.1-2025-04-14. Collapse policy: highest reported score per canonical model.

5.5%

#15 · claude-opus-4-1-20250805-thinking

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 41.7%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-1-20250805-thinking. Collapse policy: highest reported score per canonical model.

5.2%

#16 · Claude Opus 4.1

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-1-20250805. Collapse policy: highest reported score per canonical model.

4.7%

#17 · Claude Opus 4

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-1-20250805. Collapse policy: highest reported score per canonical model. Backfilled from Claude Opus 4.1 via approved benchmark identity mapping map-claude-opus-4-to-4-1.

4.7%

#18 · Claude Opus 4.7

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

4.7%

#19 · Gemini 2.5 Flash

SL · Oct 3, 2026

Source label: gemini-2.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 25%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-2.5-flash. Collapse policy: highest reported score per canonical model.

4.7%

#20 · Claude Sonnet 4

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 20.8%
Last updated: recent
Eligibility: historical_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4. Collapse policy: highest reported score per canonical model.

4.5%

#21 · Claude Sonnet 4.6

SL · Oct 3, 2026

Source label: claude-sonnet-4

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 20.8%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4. Collapse policy: highest reported score per canonical model. Backfilled from Claude Sonnet 4 via approved benchmark identity mapping map-claude-sonnet-4-6-to-4.

4.5%

#22 · Claude Sonnet 4 (Thinking)

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-thinking. Collapse policy: highest reported score per canonical model.

4.4%

#23 · Nova Premier

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 8.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: nova-premier. Collapse policy: highest reported score per canonical model.

#24 · llama4-scout

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 4.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: llama4-scout. Collapse policy: highest reported score per canonical model.

1.6%

#25 · Llama 4 Maverick

SL · Oct 3, 2026

Source label: llama4-maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 0%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: llama4-maverick. Collapse policy: highest reported score per canonical model.

1.4%

Benchmarks · /benchmarks/scale-vtb

VTB

Scale VTB leaderboard for image-grounded reasoning and multimodal understanding.

Source · Scale Labs
Version · scale-labs snapshot 2026-10-03
Scores · 25

Test details

Visible tradeoffsThis is a rubric-judged signal, so it is more structured than arena taste but still depends on the scoring rubric.

source

Scale Labs

metric

APR (%)

judge

Rubric

direction

higher better

group id

scale_vtb_current

domain

Vision understanding

What it measures vs what it misses

✓ Measures

Visual interpretation and reasoning over benchmark images and prompts.

✗ Misses

Image generation quality. Tool-use orchestration beyond the judged task.

Leaderboard · this benchmark version

#1 · muse-spark-1.1

SL · Oct 3, 2026

Source label: Muse Spark 1.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 100%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: Muse Spark 1.1. Collapse policy: highest reported score per canonical model.

44.8%

#2 · GPT-5.4

SL · Oct 3, 2026

Source label: gpt-5.4-2026-03-05 (reasoning effort = high)

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 95.8%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5.4-2026-03-05 (reasoning effort = high). Collapse policy: highest reported score per canonical model.

29.2%

#3 · Gemini 3.1 Pro Preview

SL · Oct 3, 2026

Source label: gemini-3.1-pro-preview

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 91.7%
Last updated: recent
Eligibility: preview_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-3.1-pro-preview. Collapse policy: highest reported score per canonical model.

29%

#4 · Claude Opus 4.6

SL · Oct 3, 2026

Source label: claude-opus-4-6-thinking

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 87.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-6-thinking. Collapse policy: highest reported score per canonical model.

27.5%

#5 · Gemini 3 Pro Preview

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 83.3%
Last updated: recent
Eligibility: preview_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-3-pro-preview. Collapse policy: highest reported score per canonical model.

26.9%

#6 · GPT-5

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-5-2025-08-07-thinking. Collapse policy: highest reported score per canonical model.

18.7%

#7 · GPT-5.4 mini

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

18.7%

#8 · GPT-5.4 nano

SL · Oct 3, 2026

Source label: gpt-5-2025-08-07-thinking

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 79.2%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

18.7%

#9 · o3

SL · Oct 3, 2026

Source label: o3-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 66.7%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: o3-2025-04-16. Collapse policy: highest reported score per canonical model.

13.7%

#10 · Gemini 2.5 Pro

SL · Oct 3, 2026

Source label: gemini-2.5-pro-preview-06-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 62.5%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-2.5-pro-preview-06-05. Collapse policy: highest reported score per canonical model.

11.8%

#11 · o4 mini

SL · Oct 3, 2026

Source label: o4-mini-2025-04-16

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 58.3%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: o4-mini-2025-04-16. Collapse policy: highest reported score per canonical model.

11.1%

#12 · claude-sonnet-4-5-20250929-thinking

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 54.2%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-5-20250929-thinking. Collapse policy: highest reported score per canonical model.

6.2%

#13 · Claude Sonnet 4.5

SL · Oct 3, 2026

Source label: claude-sonnet-4-5-20250929

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 50%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-5-20250929. Collapse policy: highest reported score per canonical model.

5.6%

#14 · GPT-4.1

SL · Oct 3, 2026

Source label: gpt-4.1-2025-04-14

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 45.8%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gpt-4.1-2025-04-14. Collapse policy: highest reported score per canonical model.

5.5%

#15 · claude-opus-4-1-20250805-thinking

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 41.7%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-1-20250805-thinking. Collapse policy: highest reported score per canonical model.

5.2%

#16 · Claude Opus 4.1

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: historical_model
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-opus-4-1-20250805. Collapse policy: highest reported score per canonical model.

4.7%

#17 · Claude Opus 4

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

4.7%

#18 · Claude Opus 4.7

SL · Oct 3, 2026

Source label: claude-opus-4-1-20250805

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 37.5%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

4.7%

#19 · Gemini 2.5 Flash

SL · Oct 3, 2026

Source label: gemini-2.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 25%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: gemini-2.5-flash. Collapse policy: highest reported score per canonical model.

4.7%

#20 · Claude Sonnet 4

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 20.8%
Last updated: recent
Eligibility: historical_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4. Collapse policy: highest reported score per canonical model.

4.5%

#21 · Claude Sonnet 4.6

SL · Oct 3, 2026

Source label: claude-sonnet-4

backfilledproxy backfilledBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 20.8%
Last updated: recent
Eligibility: Fallback benchmark identity is visible for context but excluded from default ranking.
Identity: benchmark proxy (0.58)

4.5%

#22 · Claude Sonnet 4 (Thinking)

SL · Oct 3, 2026

verified runtimeexact direct

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 12.5%
Last updated: recent
Eligibility: headline eligible
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: claude-sonnet-4-thinking. Collapse policy: highest reported score per canonical model.

4.4%

#23 · Nova Premier

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 8.3%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: nova-premier. Collapse policy: highest reported score per canonical model.

#24 · llama4-scout

SL · Oct 3, 2026

verified runtimeexact directBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 4.2%
Last updated: recent
Eligibility: benchmark_derived_model
Identity: exact (1.00)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: llama4-scout. Collapse policy: highest reported score per canonical model.

1.6%

#25 · Llama 4 Maverick

SL · Oct 3, 2026

Source label: llama4-maverick

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://labs.scale.com/leaderboard/vtb
Percentile: 0%
Last updated: recent
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from the public Scale Labs page for scale-vtb. Reported model configuration: llama4-maverick. Collapse policy: highest reported score per canonical model.

1.4%

VTB

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version

Loading benchmark evidence.

VTB

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version