Benchmark Almanac

CypherBench

Claude 3.5 Sonnet 61.6% leads since December 2024
Not in model cards

Models translate questions about Wikidata into Cypher queries. The scorer executes each query and compares its result with the expected answer.

The CypherGuard team ran all 2,348 test questions across seven graphs. The theorem team also ran the full test with Claude Haiku 4.5.

Released
December 2024
Kind
Data science and SQL
Level
University
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Claude 3.5 Sonnet #158 on the capability index
Last model added
September 2024
Scores refreshed
November 20th, 2026
12
Models scored in its paper

Claude 3.5 Sonnet leads with 61.6% among models released by September 2024

Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude 3.5 Sonnet 61.6%
  2. 2 GPT-4o Aug 2024 60.2%
  3. 3 Qwen2.5-72B 41.9%
  4. 4 Gemini 1.5 Pro May 2024 40%
  5. 5 Llama 3.1-70B 38.8%
  6. 6 Yi-Large 33.8%
  7. 7 GPT-4o mini 31.4%
  8. 8 Gemini 1.5 Flash May 2024 25.3%
  9. 9 Mixtral 8x7B 19.2%
  10. 10 Llama 3.1-8B 18.8%
  11. 11 Gemma 2 9B 18.6%
  12. 12 Llama 3.2 3B 11.2%

Source: The CypherBench paper

#ModelScoreSettingDateSource
1 Claude 3.5 Sonnet 61.6% Jun 2024 Official leaderboard
2 GPT-4o Aug 2024 60.2% Aug 2024 Official leaderboard
3 Qwen2.5-72B 41.9% Sep 2024 Official leaderboard
4 Gemini 1.5 Pro May 2024 40% May 2024 Official leaderboard
5 Llama 3.1-70B 38.8% Jul 2024 Official leaderboard
6 Yi-Large 33.8% May 2024 Official leaderboard
7 GPT-4o mini 31.4% Jul 2024 Official leaderboard
8 Gemini 1.5 Flash May 2024 25.3% May 2024 Official leaderboard
9 Mixtral 8x7B 19.2% Dec 2023 Official leaderboard
10 Llama 3.1-8B 18.8% Jul 2024 Official leaderboard
11 Gemma 2 9B 18.6% Jun 2024 Official leaderboard
12 Llama 3.2 3B 11.2% Sep 2024 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s paper: PSJS and Executable percentage. Each ranks models on one part of the test. Claude 3.5 Sonnet leads both. Arrows show how far a model moved from its overall rank among the models we track.

PSJS

12 models we track have a score here.

  1. 1 Claude 3.5 Sonnet 80.8%
  2. 2 GPT-4o Aug 2024 76.9%
  3. 3 Gemini 1.5 Pro May 2024 ↑1 57.7%
  4. 4 Qwen2.5-72B ↓1 56.4%
  5. 5 Llama 3.1-70B 54.8%
  6. 6 Yi-Large 47.2%
  7. 7 GPT-4o mini 45.9%
  8. 8 Gemini 1.5 Flash May 2024 41.5%
  9. 9 Mixtral 8x7B 37%
  10. 10 Llama 3.1-8B 31%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Benchmark paper Not tested since Dec 2024

Source: The CypherBench paper

Scores over time

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

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Claude 3.5 Sonnet Gemini 1.5 Pro Yi-Large Mixtral 8x7B 0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) Claude 3.5 Sonnet Gemini 1.5 Pro

Source: Official CypherBench leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Coding

All 523 in Coding
  • SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
  • LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
  • SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
  • DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
  • SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases. Seed2.1 Pro 59.8%
  • Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in September 2024, for Llama 3.2 3B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (12 entries).
  • Pass Models scored 12 models we track.
  • Warning Newest score The newest came 2 years ago, in September 2024, for Llama 3.2 3B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude 3.5 Sonnet (#158).
  • Pass Headroom The best is 61.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on aclanthology.org: 12 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check