# Vals CaseLaw v2 (CaseLaw v2)

> Vals AI private question-answer benchmark over Canadian court cases.

Canonical page: https://benchlm.ai/benchmarks/valscaselawv2

- Category: [Knowledge](/knowledge)
- Last updated: May 4, 2026

## About CaseLaw v2

- Year: 2026
- Tasks: Canadian case-law question answering
- Format: Accuracy score
- Difficulty: Professional legal retrieval and reasoning
- Paper: [CaseLaw v2](https://www.vals.ai/benchmarks/case_law_v2)

Vals marks CaseLaw v2 as archived. BenchLM mirrors the public leaderboard as display-only historical legal-domain context.

CaseLaw v2 is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (54 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Grok 4.3](/models/grok-4-3) | xAI | 79.31% |
| 2 | [GPT-5.1](/models/gpt-5-1) | OpenAI | 73.42% |
| 3 | [GPT-4.1](/models/gpt-4-1) | OpenAI | 69.88% |
| 4 | [GPT-5 mini](/models/gpt-5-mini) | OpenAI | 68.49% |
| 5 | [Claude Opus 4.7](/models/claude-opus-4-7) | Anthropic | 68.38% |
| 6 | [GPT-5](https://www.vals.ai/models/openai_gpt-5-2025-08-07) | OpenAI | 66.45% |
| 7 | [GPT-5.5](/models/gpt-5-5) | OpenAI | 66.24% |
| 8 | [GPT-5.2](/models/gpt-5-2) | OpenAI | 66.02% |
| 9 | [Grok 4 0709](https://www.vals.ai/models/grok_grok-4-0709) | xAI | 65.81% |
| 10 | [Grok 4 Fast (Reasoning)](/models/grok-4-fast-reasoning) | xAI | 65.70% |
| 11 | [Kimi K2 Thinking](https://www.vals.ai/models/kimi_kimi-k2-thinking) | Moonshot AI | 65.70% |
| 12 | [Gemini 3.1 Pro Preview](https://www.vals.ai/models/google_gemini-3.1-pro-preview) | Google | 64.84% |
| 13 | [Command A 03 2025](https://www.vals.ai/models/cohere_command-a-03-2025) | Cohere | 64.52% |
| 14 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Anthropic | 63.99% |
| 15 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | 63.88% |
| 16 | [GPT-5.4](/models/gpt-5-4) | OpenAI | 63.77% |
| 17 | [Muse Spark](/models/muse-spark) | Meta | 63.13% |
| 18 | [Claude Opus 4.5 Thinking](/models/claude-opus-4-5-thinking) | Anthropic | 62.59% |
| 19 | [Claude Sonnet 4.5 Thinking](/models/claude-sonnet-4-5-thinking) | Anthropic | 62.16% |
| 20 | [Claude Opus 4.6 (Adaptive)](/models/claude-opus-4-6-thinking) | Anthropic | 62.06% |
| 21 | [Mistral Large 2512](https://www.vals.ai/models/mistralai_mistral-large-2512) | Mistral | 61.41% |
| 22 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | 61.20% |
| 23 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | 60.88% |
| 24 | [Grok 4.1 Fast (Reasoning)](/models/grok-4-1-fast-reasoning) | xAI | 60.45% |
| 25 | [GPT-4o](/models/gpt-4o) | OpenAI | 59.70% |
| 26 | [Qwen3.5 Plus Thinking](https://www.vals.ai/models/alibaba_qwen3.5-plus-thinking) | Alibaba | 59.70% |
| 27 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | 59.38% |
| 28 | [Kimi K2.5 Thinking](https://www.vals.ai/models/kimi_kimi-k2.5-thinking) | Moonshot AI | 58.73% |
| 29 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | 57.88% |
| 30 | [Claude Haiku 4.5 Thinking](/models/claude-haiku-4-5-thinking) | Anthropic | 56.48% |
| 31 | [Qwen3.5 Flash](/models/qwen3-5-flash) | Alibaba | 55.95% |
| 32 | [MiniMax M2.1](https://www.vals.ai/models/minimax_MiniMax-M2.1) | MiniMax | 55.84% |
| 33 | [Gemini 3 Flash Preview](https://www.vals.ai/models/google_gemini-3-flash-preview) | Google | 55.84% |
| 34 | [DeepSeek V3p2 Thinking](https://www.vals.ai/models/fireworks_deepseek-v3p2-thinking) | Fireworks | 55.41% |
| 35 | [Gemini 3.1 Flash Lite Preview](https://www.vals.ai/models/google_gemini-3.1-flash-lite-preview) | Google | 54.98% |
| 36 | [Qwen3 Max](/models/qwen3-max) | Alibaba | 54.98% |
| 37 | [GLM-4.7](/models/glm-4-7) | Z.AI | 54.88% |
| 38 | [Grok 4.20 0309 Reasoning](https://www.vals.ai/models/grok_grok-4.20-0309-reasoning) | xAI | 54.45% |
| 39 | [DeepSeek V3p1](https://www.vals.ai/models/fireworks_deepseek-v3p1) | Fireworks | 53.91% |
| 40 | [MiniMax M2.5](/models/minimax-m2-5) | MiniMax | 53.48% |
| 41 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | 53.16% |
| 42 | [Gemini 3 Pro Preview](https://www.vals.ai/models/google_gemini-3-pro-preview) | Google | 53.05% |
| 43 | [Gemma 4 31b It](https://www.vals.ai/models/google_gemma-4-31b-it) | Google | 52.63% |
| 44 | [GPT-5 nano](/models/gpt-5-nano) | OpenAI | 52.63% |
| 45 | [GLM 5 Thinking](https://www.vals.ai/models/zai_glm-5-thinking) | zAI | 52.52% |
| 46 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | 51.88% |
| 47 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | 51.66% |
| 48 | [GLM-5.1](/models/glm-5-1) | Z.AI | 51.55% |
| 49 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | 51.45% |
| 50 | [GPT-OSS 120B](/models/gpt-oss-120b) | OpenAI | 48.77% |
| 51 | [Qwen 3.6 Max (preview)](/models/qwen3-6-max-preview) | Alibaba | 47.91% |
| 52 | [Qwen3 Max](/models/qwen3-max) | Alibaba | 47.48% |
| 53 | [Mistral Medium 3.5](https://www.vals.ai/models/mistralai_mistral-medium-3.5) | Mistral | 44.16% |
| 54 | [GPT-OSS 20B](/models/gpt-oss-20b) | OpenAI | 43.84% |

## FAQ

### What does CaseLaw v2 measure?

Vals AI private question-answer benchmark over Canadian court cases.

### Which model leads the published CaseLaw v2 snapshot?

Grok 4.3 currently leads the published CaseLaw v2 snapshot with a score of 79.31%.

### How many models are evaluated on CaseLaw v2?

The May 4, 2026 contains 54 AI models.

### Does CaseLaw v2 affect BenchLM's overall score?

Not directly. CaseLaw v2 is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.