trace: verified ✓ 23 steps
confidence 0.86 · calibrated
step 7/23 · planning
P(claim) 0.003 → flag
Open reasoning benchmarks

Measure how models think,
not just what they answer.

ReasonEval is an open suite of trace-level reasoning benchmarks — math, code, logic and long-horizon planning — graded on process quality, calibration and trust, not final answers alone.

6 benchmark suites· 8,412 problems· 214k graded traces· Updated Sept 2026
8,412
curated problems
27
frontier models tracked
214k
reasoning traces graded
93.4%
dual-judge agreement
The ReasonBench suite

Six suites, one question:
can the model actually reason?

Every suite grades the full reasoning trace — not just the final answer — under adversarial perturbation and calibration probes.

ReasonBench · Math

Competition and formal-verified mathematics. Answers are checked by CAS verifiers; proofs are graded step-by-step.

1,842 problemsCAS-verified

ReasonBench · Code

Algorithmic and agentic coding tasks scored against executable test suites in sandboxed runtimes.

1,655 problemssandboxed

ReasonBench · Logic

Formal logic, constraint satisfaction and proof completion with machine-checkable ground truth.

1,208 problemscheckable

ReasonBench · MultiStep

Long-horizon planning tasks requiring 15+ chained tool calls, state tracking and recovery from dead ends.

1,116 problems15+ steps

ReasonBench · Calib

Epistemic calibration: separable confidence on knowns, honest uncertainty on unknowns and traps.

964 problemsECE-scored

ReasonBench · Agent

Multi-turn agentic evals in sandboxed environments — browsing, file systems and APIs with replayable logs.

1,627 problemsreplayable
Leaderboard · Sept 2026

The current reasoning frontier

Composite R-score across all six suites, weighted 40/25/20/15 for correctness, process, calibration and robustness.

# Model Organization Overall R-score Math Code
1 Claude Opus 4.5 Anthropic 89.2
91.4 90.1
2 GPT-5.2 OpenAI 88.6
92.0 88.9
3 Gemini 3 Pro Google 87.9
89.8 87.2
5 DeepSeek-R2Open DeepSeek 84.7
87.9 86.4
6 GLM-5Open Z.ai 84.0
85.2 85.8

Showing 5 of 27 tracked models. View the full leaderboard →

Why it's different

Grading the reasoning, not the result

Trace-level grading

Every step of the chain-of-thought is scored against rubrics and formal verifiers — lucky guesses don't pass, sound reasoning that trips on arithmetic doesn't crash.

Adversarial perturbation

Each problem ships with rewritten variants — reordered facts, changed constants, added distractors — to detect memorization and score genuine robustness.

Calibrated trust

We reward models for saying "I don't know" when they shouldn't know. Overconfidence is penalized; honest uncertainty is scored as a feature.

Start evaluating in minutes

Point the CLI at any OpenAI-compatible endpoint and get a full trace-graded report — per-step scores, calibration curves and failure taxonomies.

Get started
terminal
$ pip install reasoneval
$ reasoneval run --model "my-model" --suite all
▸ math        87.9  (±1.2)   1,842 problems
▸ code        86.4  (±1.4)   1,655 problems
▸ logic       82.1  (±1.6)   1,208 problems
▸ R-score     84.7  (±0.9)   ✓ report.html written