Skip to main content
benchlm.ai

Tool-Agent-User Benchmark (TAU-bench)

TAU-bench evaluates AI agents in realistic enterprise scenarios requiring multi-turn tool use, database interactions, and policy adherence. It tests across retail and airline domains, measuring an agent's ability to reliably complete customer service tasks while following complex business rules.

How BenchLM shows TAU-bench right now

BenchLM is tracking TAU-bench in the local dataset, but exact-source verification records for these rows are still being attached. To avoid a blank benchmark page, BenchLM shows the current tracked rows below as a display-only reference table.

These tracked rows are useful for inspection and spot-checking, but until exact-source attachments are completed they should not be treated as fully verified public benchmark rows.

38 tracked modelsLocal tracked rowsAwaiting exact-source attachmentsDisplay only

Tracked score on TAU-bench — July 6, 2026

BenchLM mirrors the published tracked score view for TAU-bench. Claude Mythos 5 leads the public snapshot at 89.2% , followed by Claude Fable 5 (89.2%) and Claude Sonnet 4.6 (87.5%). BenchLM does not use these results to rank models overall.

1

Claude Mythos 5

Anthropic

claude-mythos-5

89.2%
Overall —
2

Claude Fable 5

Anthropic

claude-fable-5

89.2%
Overall —
3

Claude Sonnet 4.6

Anthropic

claude-sonnet-4-6

87.5%
Overall —
38 modelsAgenticRefreshingDisplay onlyUpdated July 6, 2026

The published TAU-bench snapshot is tightly clustered at the top: Claude Mythos 5 sits at 89.2%, while the third row is only 1.7 points behind. The broader top-10 spread is 9.1 points, so many of the published scores sit in a relatively narrow band.

38 models have been evaluated on TAU-bench. The benchmark falls in the Agentic category. This category carries a 22% weight in BenchLM.ai's overall scoring system. TAU-bench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About TAU-bench

Year

2024

Tasks

680

Public benchmark source

BenchLM freshness & provenance

Version

TAU-bench 2024

Refresh cadence

Annual

Staleness state

Refreshing

Question availability

Public benchmark set

RefreshingDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Tracked score table (38 models)

1
Claude Mythos 5claude-mythos-5
89.2%
2
Claude Fable 5claude-fable-5
89.2%
3
Claude Sonnet 4.6claude-sonnet-4-6
87.5%
4
Claude Sonnet 4.5claude-sonnet-4-5
86.2%
5
Claude Opus 4.6claude-opus-4-6
84.8%
6
GLM-5 (Reasoning)glm-5-reasoning
83.4%
7
Claude 4.1 Opusclaude-4-1-opus
82.4%
8
GLM-5glm-5
82.1%
9
Grok 4.20 Multi-agentgrok-4-20-multi-agent-beta
80.5%
10
GPT-5.4 Progpt-5-4-pro
80.1%
11
GLM-4.5glm-4-5
79.7%
12
Grok 4.20grok-4-20-beta
78.9%
13
GPT-5.4gpt-5-4
78.3%
14
Qwen3.5 397B (Reasoning)qwen3-5-397b-reasoning
78.2%
15
GPT-5.3 Codexgpt-5-3-codex
77.8%
16
Qwen3.5 397Bqwen3-5-397b
77.5%
17
Qwen3.6 Plusqwen3-6-plus
76.8%
18
Gemini 3.1 Progemini-3-1-pro
76.5%
19
Step 3.5 Flashstep-3-5-flash
76.2%
20
Gemini 3 Progemini-3-pro
75.3%
21
GPT-5.2gpt-5-2
75.1%
22
Llama 4 Behemothllama-4-behemoth
74.8%
23
Grok 4.1grok-4-1
74.6%
24
GPT-5.1gpt-5-1
74.2%
25
Kimi K2.5kimi-k2-5
74.2%
26
DeepSeek V3.2 (Thinking)deepseek-v3-2-thinking
73.5%
27
GPT-5.3 Instantgpt-5-3-instant
73.4%
28
Grok 4.1 Fastgrok-4-1-fast
72.8%
29
MiniMax M2.7minimax-m2-7
72.8%
30
o4-mini (high)o4-mini-high
71.8%
31
Gemini 3 Flashgemini-3-flash
71.5%
32
DeepSeek V3.2deepseek-v3-2
71.3%
33
Mistral Large 3mistral-large-3
70.2%
34
MiMo-V2-Promimo-v2-pro
69.2%
35
Llama 4 Maverickllama-4-maverick
68.5%
36
Qwen3.5 Flashqwen3-5-flash
66.4%
37
Mistral Small 4mistral-small-4
65.8%
38
Llama 4 Scoutllama-4-scout
62.3%

FAQ

What does TAU-bench measure?

TAU-bench evaluates AI agents in realistic enterprise scenarios requiring multi-turn tool use, database interactions, and policy adherence. It tests across retail and airline domains, measuring an agent's ability to reliably complete customer service tasks while following complex business rules.

Which model leads the published TAU-bench snapshot?

Claude Mythos 5 currently leads the published TAU-bench snapshot with 89.2% tracked score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on TAU-bench?

38 AI models are included in BenchLM's mirrored TAU-bench snapshot, based on the public leaderboard captured on July 6, 2026.

Last updated: July 6, 2026 · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.