Skip to content

Benchmark

tau3-bench v1 banking-knowledge pass^3

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
25
Scores recorded
26
Sources
1
IQ weight
1.62

Score distribution

18.3%
Mean
18.3%
Median
11.1%
Std dev
5.7%
10th percentile
34.5%
90th percentile
2.6%
Min
2.6%39.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.62 for chat models, measured over 25 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Qwen3.8 Maxsubmission-qwen3-8-max_sierra_2026-08-04, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs39.7%
2Grok 4.5submission-grok-4-5_sierra_2026-08-04, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs35.3%
3Claude Opus 5submission-claude-opus-5_sierra_2026-08-04, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs34.5%
4GPT 5.5submission-gpt-5-5_sierra_2026-05-05, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs33.2%
5GPT-5.6 Solsubmission-gpt-5-6-sol_sierra_2026-08-04, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs32%
6Claude Fable 5submission-claude-fable-5_sierra_2026-08-04, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs30.2%
7Claude Opus 4.7submission-claude-opus-4-7_sierra_2026-05-05, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs27.1%
8Claude Opus 4.8submission-claude-opus-4-8_sierra_2026-08-04, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs26.3%
9GPT 5.4submission-gpt-5-4_sierra_2026-03-25, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs24.5%
10Muse Spark 1.1submission-muse-spark-1-1_sierra_2026-08-04, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs24%
11GPT 5.2submission-gpt-5-2_sierra_2026-02-26, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs20.6%
12Kimi K3submission-kimi-k3_sierra_2026-08-04, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs20.1%
13GLM 5.2submission-glm-5-2_sierra_2026-08-04, usersim-gpt-5.2, effort-xhigh, retrieval-alltools· Sierra tau-bench Leaderboard data docs18.3%
14Claude Opus 4.6submission-claude-opus-4-6_sierra_2026-05-05, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs14.2%
15Claude Opus 4.5submission-claude-opus-4-5_sierra_2026-02-26, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs13.7%
16Inkling FP4submission-inkling_sierra_2026-08-04, usersim-gpt-5.2, effort-max, retrieval-alltools· Sierra tau-bench Leaderboard data docs13.4%
17Claude Sonnet 4.5submission-claude-sonnet-4-5_sierra_2026-02-26, usersim-gpt-5.2, effort-enabled, retrieval-terminal· Sierra tau-bench Leaderboard data docs12.6%
18Gemini 3.1 Pro Previewsubmission-gemini-3-1-pro-preview_sierra_2026-05-05, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs11.1%
19Gemini 3 Flashsubmission-gemini-3-flash_sierra_2026-03-02, usersim-gpt-5.2, effort-high, retrieval-terminal· Sierra tau-bench Leaderboard data docs10.3%
20Gemini 3 Pro Previewsubmission-gemini-3-pro_sierra_2026-03-02, usersim-gpt-5.2, effort-high, retrieval-terminal· Sierra tau-bench Leaderboard data docs5.7%
21GLM 5submission-glm-5-think_sierra_2026-03-02, usersim-gpt-5.2, effort-enabled, retrieval-text-emb-3-large· Sierra tau-bench Leaderboard data docs5.7%
22Grok 4 Fastsubmission-grok-4-fast_sierra_2026-05-05, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs5.7%
23Grok 4.1 Fastsubmission-grok-4-1-fast_sierra_2026-05-05, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs5.7%
24Qwen3.5 397B A17Bsubmission-qwen3.5-397b-a17b-think_sierra_2026-03-02, usersim-gpt-5.2, effort-enabled, retrieval-text-emb-3-large· Sierra tau-bench Leaderboard data docs5.4%
25GPT 5.2submission-gpt-5-2-none_sierra_2026-02-26, usersim-gpt-5.2, effort-none, retrieval-qwen-embeddings· Sierra tau-bench Leaderboard data docs5.2%
26Gemini 2.5 Prosubmission-gemini-2-5-pro_sierra_2026-05-05, usersim-gpt-5.2, effort-high, retrieval-alltools· Sierra tau-bench Leaderboard data docs2.6%

Data credits: Sierra Research tau2-bench leaderboard submissions (github.com/sierra-research/tau2-bench), MIT (licence)