BenchmarkActivecategory · agentic
τ²-bench
dual-control tool-agent-user interaction (telecom, retail, airline)
quality51
Updated 9 h ago · first seen 11 Sept 2026
bench_01M293SPH15XRN6880KKHSKM5P
- Metric
- pass^1 · %
- Direction
- Higher is better
- Results
- 440
- Leader
- Z.ai GLM 5.2 99.12%
Score history · gemini-3-pro 1 row
Not enough history to chart — a single observation (87.13% on 11 Sept 2026). Rows under different configurations count separately; the list below shows each one.
- 87.13%aa_slug=gemini-3-pro · evaluator=Artificial Analysis · index_version=4.311 Sept 2026
Leaderboard 440 current results
Select models with +, then open Compare.
| # | Model | Score | Config | Evaluated | Source | Actions |
|---|---|---|---|---|---|---|
| #401llama-3-1-nemotron-ultra-253b-v1-reasoningNVIDIA | 11.4% | aa_slug=llama-3-1-nemotron-ultra-253b-v1-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #402apertus-8b-instructSwiss AI Initiative | 11.4% | aa_slug=apertus-8b-instruct · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #403deepseek-r1-0120DeepSeek | 11.4% | aa_slug=deepseek-r1-0120 · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #404lfm2-24b-a2bLiquid AI | 11.11% | aa_slug=lfm2-24b-a2b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #405LFM2.5-1.2B-InstructLiquid AI | 10.82% | aa_slug=lfm2-5-1-2b-instruct · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #406Gemma 3 12BGoogle | 10.82% | aa_slug=gemma-3-12b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #407Gemma 3 27BGoogle | 10.53% | aa_slug=gemma-3-27b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #408lfm2-8b-a1bLiquid AI | 10.53% | aa_slug=lfm2-8b-a1b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #409granite-3-3-8b-instructIBM | 10.53% | aa_slug=granite-3-3-8b-instruct · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #410gemma-3-1bGoogle | 10.53% | aa_slug=gemma-3-1b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #411qwen3-30b-a3b-2507Alibaba Group | 10.23% | aa_slug=qwen3-30b-a3b-2507 · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #412gemma-3-270mGoogle | 9.06% | aa_slug=gemma-3-270m · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #413LFM2.5-VL-1.6BLiquid AI | 8.48% | aa_slug=lfm2-5-vl-1-6b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #414phi-4-miniMicrosoft | 8.19% | aa_slug=phi-4-mini · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #415Gemma 3 4BGoogle | 4.97% | aa_slug=gemma-3-4b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #416gemma-3n-e4bGoogle | 4.97% | aa_slug=gemma-3n-e4b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #417exaone-4-0-32bLG AI Research | 4.09% | aa_slug=exaone-4-0-32b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #418Kimi-Linear-48B-A3B-InstructMoonshot AI | 0% | aa_slug=kimi-linear-48b-a3b-instruct · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #419ernie-4-5-300b-a47bBaidu | 0% | aa_slug=ernie-4-5-300b-a47b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #420phi-3-miniMicrosoft | 0% | aa_slug=phi-3-mini · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #421Mistral 7BMistral AI | 0% | aa_slug=mistral-7b-instruct · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #422ring-flash-2-0inclusionAI | 0% | aa_slug=ring-flash-2-0 · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #423gpt-5-chatgptOpenAI | 0% | aa_slug=gpt-5-chatgpt · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #424Phi 4Microsoft | 0% | aa_slug=phi-4 · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #425llama-3-instruct-70bMeta AI | 0% | aa_slug=llama-3-instruct-70b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #426olmo-2-32bAllen Institute for AI | 0% | aa_slug=olmo-2-32b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #427olmo-2-7bAllen Institute for AI | 0% | aa_slug=olmo-2-7b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #428gemma-3n-e2bGoogle | 0% | aa_slug=gemma-3n-e2b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #429olmo-3-1-32b-thinkAllen Institute for AI | 0% | aa_slug=olmo-3-1-32b-think · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #430deepseek-v3-2-specialeDeepSeek | 0% | aa_slug=deepseek-v3-2-speciale · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #431olmo-3-32b-thinkAllen Institute for AI | 0% | aa_slug=olmo-3-32b-think · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #432sarvam-m-reasoningSarvam | 0% | aa_slug=sarvam-m-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #433Olmo-3-7B-ThinkAllen Institute for AI | 0% | aa_slug=olmo-3-7b-think · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #434Llama-3.2-1BMeta AI | 0% | aa_slug=llama-3-2-instruct-1b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #435tiny-aya-globalCohere | 0% | aa_slug=tiny-aya-global · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #436Molmo2-8BAllen Institute for AI | 0% | aa_slug=molmo2-8b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #437llama-3-instruct-8bMeta AI | 0% | aa_slug=llama-3-instruct-8b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #438DeepSeek-R1-0528-Qwen3-8BDeepSeek | 0% | aa_slug=deepseek-r1-qwen3-8b · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #439Reka Flash 3rekaai | 0% | aa_slug=reka-flash-3 · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #440molmo-7b-dAllen Institute for AI | 0% | aa_slug=molmo-7b-d · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). The bar is relative to the best score on this page. Results with different configs are not directly comparable — see methodology.
Definition
- Category
- agentic
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 9 h agomedium
- Task
- dual-control tool-agent-user interaction (telecom, retail, airline)
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 9 h agomedium
- Metric
- pass^1 · %
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 9 h agomedium
- Direction
- Higher is better
- Paper
- https://arxiv.org/abs/2506.07982
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 9 h agomedium
Each value shows its source, tier and observation time. Missing rows mean no source stated them. How results are recorded →
- Evaluated models
- grok-3-mini-reasoning, gemma-4-12B, grok-4-3-low, GPT-5.1-Codex Mini, Qwen3 Coder Next, KAT-Coder-Pro V2, nova-2-0-lite-reasoning-medium, lfm2-24b-a2b +432(440 total)
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history
5 claims · 5 properties
Categorycategory1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| agentic | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Metricmetric1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| pass^1 | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Paperpaper1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://arxiv.org/abs/2506.07982 | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Tasktask1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| dual-control tool-agent-user interaction (telecom, retail, airline) | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Unitunit1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| % | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Artificial Analysis | artificialanalysis.ai/leaderboards/models | leaderboard | T2· Quality secondary | 8 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.