Benchmarks · /benchmarks/vals-ai-hlab

Harvey's Legal Agent Benchmark

Name: Harvey's Legal Agent Benchmark
Creator: Vals AI

Harvey's Legal Agent Benchmark result as reported by Vals AI.

Source · Vals AI
Version · vals-ai snapshot 2026-06-24
Scores · 14

Test details

Verified but agingThis is an objective signal, so it is mainly about measurable task performance rather than public taste.

source

Vals AI

metric

Accuracy (%)

judge

Objective

direction

higher better

group id

vals_hlab_current

domain

Professional reasoning

What it measures vs what it misses

✓ Measures

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

✗ Misses

Adjacent skills outside the benchmark task mix, latency, and cost.

Why this countsCompleting legal work with documents, spreadsheets, presentations, and file-system tools.Same-test ruleThis percentile only compares models inside the exact benchmark/version group shown here. It is not a universal score.What it missesAdjacent skills outside the benchmark task mix, latency, and cost.

Leaderboard · this benchmark version

#1 · Claude Fable 5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-fable-5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 100%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

11.3%7% - 15.5%

#2 · Claude Opus 4.8

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-8

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 92.3%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

9.6%5.4% - 13.8%

#3 · glm-5.2-max

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 84.6%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

7.1%3.2% - 11%

#4 · Claude Sonnet 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-6

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 76.9%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

5%1.8% - 8.2%

#5 · minimax-m3

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 69.2%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: MiniMax.

4.2%0.9% - 7.4%

#6 · deepseek-v4-pro

VALS-AI · Jun 17, 2026

Source label: deepseek/deepseek-v4-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 61.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: DeepSeek.

3.8%0.9% - 6.6%

#7 · GPT-5.5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 61.5%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

3.8%1.4% - 6.1%

#8 · Gemini 3.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 46.2%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Google.

2.5%0.9% - 4.1%

#9 · kimi-k2.6

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 38.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Moonshot AI.

1.7%0% - 3.3%

#10 · Qwen3.7 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 38.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Alibaba.

1.7%0% - 3.3%

#11 · Grok 4.3

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 23.1%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: xAI.

0.4%0.4% - 0.4%

#12 · glm-5.1

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

0%0% - 0%

#13 · qwen3.7-plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Alibaba.

0%0% - 0%

#14 · GPT-5.4

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-2026-03-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

0%0% - 0%

Benchmarks · /benchmarks/vals-ai-hlab

Harvey's Legal Agent Benchmark

Harvey's Legal Agent Benchmark result as reported by Vals AI.

Source · Vals AI
Version · vals-ai snapshot 2026-06-24
Scores · 14

Test details

Verified but agingThis is an objective signal, so it is mainly about measurable task performance rather than public taste.

source

Vals AI

metric

Accuracy (%)

judge

Objective

direction

higher better

group id

vals_hlab_current

domain

Professional reasoning

What it measures vs what it misses

✓ Measures

Completing legal work with documents, spreadsheets, presentations, and file-system tools.

✗ Misses

Adjacent skills outside the benchmark task mix, latency, and cost.

Leaderboard · this benchmark version

#1 · Claude Fable 5

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-fable-5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 100%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

11.3%7% - 15.5%

#2 · Claude Opus 4.8

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-opus-4-8

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 92.3%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

9.6%5.4% - 13.8%

#3 · glm-5.2-max

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.2

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 84.6%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

7.1%3.2% - 11%

#4 · Claude Sonnet 4.6

VALS-AI · Jun 17, 2026

Source label: anthropic/claude-sonnet-4-6

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 76.9%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Anthropic.

5%1.8% - 8.2%

#5 · minimax-m3

VALS-AI · Jun 17, 2026

Source label: minimax/MiniMax-M3

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 69.2%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: MiniMax.

4.2%0.9% - 7.4%

#6 · deepseek-v4-pro

VALS-AI · Jun 17, 2026

Source label: deepseek/deepseek-v4-pro

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 61.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: DeepSeek.

3.8%0.9% - 6.6%

#7 · GPT-5.5

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.5

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 61.5%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

3.8%1.4% - 6.1%

#8 · Gemini 3.5 Flash

VALS-AI · Jun 17, 2026

Source label: google/gemini-3.5-flash

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 46.2%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Google.

2.5%0.9% - 4.1%

#9 · kimi-k2.6

VALS-AI · Jun 17, 2026

Source label: kimi/kimi-k2.6

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 38.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Moonshot AI.

1.7%0% - 3.3%

#10 · Qwen3.7 Max

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-max

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 38.5%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Alibaba.

1.7%0% - 3.3%

#11 · Grok 4.3

VALS-AI · Jun 17, 2026

Source label: grok/grok-4.3

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 23.1%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: xAI.

0.4%0.4% - 0.4%

#12 · glm-5.1

VALS-AI · Jun 17, 2026

Source label: zai/glm-5.1

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Zhipu AI.

0%0% - 0%

#13 · qwen3.7-plus

VALS-AI · Jun 17, 2026

Source label: alibaba/qwen3.7-plus

verified runtimeexact aliasBackground only

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: benchmark_derived_model
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: Alibaba.

0%0% - 0%

#14 · GPT-5.4

VALS-AI · Jun 17, 2026

Source label: openai/gpt-5.4-2026-03-05

verified runtimeexact alias

Raw row drilldownsource, percentile, eligibility

Source URL: https://www.vals.ai/benchmarks/hlab
Percentile: 15.4%
Last updated: aging
Eligibility: headline eligible
Identity: provider alias (0.92)

Parsed from Vals AI BenchmarkView overall scores. Vals slug: hlab; provider: OpenAI.

0%0% - 0%

Harvey's Legal Agent Benchmark

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version

Loading benchmark evidence.

Harvey's Legal Agent Benchmark

Test details

What it measures vs what it misses

✓ Measures

✗ Misses

Leaderboard · this benchmark version