Agents · Benchmark profile

τ²-Bench Telecom

Dual-control conversational agent benchmark for telecom support where agent and user must coordinate tool actions.

Ranking weighted2Active90 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on τ²-Bench Telecom

Leaderboard · 90 models

GLM-5.2 leads at 99.1%, followed by Claude Fable 5 (98.5%) and GLM-5-Turbo (98.5%).

Ranking · 1.4%Agents2

#1Open

GLM-5.2

Z.AI

glm-5-2

99.1%

Overall #11 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

98.5%

Overall #2 · Context 1M tokens

#3Closed

GLM-5-Turbo

Z.AI

glm-5-turbo

98.5%

Overall #28 · Context 200K tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

Z.AIAnthropicxAIAlibaba CloudDeepSeekMoonshot AIGoogleMiniMaxXiaomi

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
GLM-5.2	#1	Z.AI	99.1%
Claude Fable 5	#2	Anthropic	98.5%
GLM-5-Turbo	#3	Z.AI	98.5%
GLM-5V-Turbo	#4	Z.AI	98.5%
GLM-5	#5	Z.AI	98.2%
GLM-5.1	#6	Z.AI	97.7%
Grok 4.3	#7	xAI	97.7%
Qwen3.6 Plus	#8	Alibaba Cloud	97.7%
DeepSeek V4 Pro	#9	DeepSeek	96.2%
GLM-4.7	#10	Z.AI	95.9%
Kimi K2.5	#11	Moonshot AI	95.9%
Kimi K2.6	#12	Moonshot AI	95.9%
Qwen3.6 Max	#13	Alibaba Cloud	95.9%
Gemini 3.1 Pro Preview	#14	Google	95.6%
Qwen3.5 397B	#15	Alibaba Cloud	95.6%
Gemini 3.5 Flash	#16	Google	95.3%
MiniMax M2.5	#17	MiniMax	95.3%
DeepSeek V4 Flash	#18	DeepSeek	95.0%
MiMo-V2-Flash	#19	Xiaomi	95.0%
MiMo-V2-Pro	#20	Xiaomi	95.0%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
GLM-5.2	#1	Z.AI	99.1%
Claude Fable 5	#2	Anthropic	98.5%
GLM-5-Turbo	#3	Z.AI	98.5%
GLM-5V-Turbo	#4	Z.AI	98.5%
GLM-5	#5	Z.AI	98.2%
GLM-5.1	#6	Z.AI	97.7%
Grok 4.3	#7	xAI	97.7%
Qwen3.6 Plus	#8	Alibaba Cloud	97.7%
DeepSeek V4 Pro	#9	DeepSeek	96.2%
GLM-4.7	#10	Z.AI	95.9%
Kimi K2.5	#11	Moonshot AI	95.9%
Kimi K2.6	#12	Moonshot AI	95.9%
Qwen3.6 Max	#13	Alibaba Cloud	95.9%
Gemini 3.1 Pro Preview	#14	Google	95.6%
Qwen3.5 397B	#15	Alibaba Cloud	95.6%
Gemini 3.5 Flash	#16	Google	95.3%
MiniMax M2.5	#17	MiniMax	95.3%
DeepSeek V4 Flash	#18	DeepSeek	95.0%
MiMo-V2-Flash	#19	Xiaomi	95.0%
MiMo-V2-Pro	#20	Xiaomi	95.0%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	1	98.5%
Z.AI	6	98.0%
xAI	1	97.7%
Alibaba Cloud	3	96.4%
Moonshot AI	2	95.9%
DeepSeek	2	95.6%
Google	2	95.5%
MiniMax	1	95.3%
Xiaomi	2	95.0%

Benchmark score table (90 models)

One best score per model · higher is better

τ²-Bench Telecom leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	GLM-5.2 glm-5-2	Z.AI	open	99.1%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	98.5%
#3	GLM-5-Turbo glm-5-turbo	Z.AI	closed	98.5%
#4	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	98.5%
#5	GLM-5 glm-5	Z.AI	open	98.2%
#6	GLM-5.1 glm-5-1	Z.AI	open	97.7%
#7	Grok 4.3 grok-4.3	xAI	closed	97.7%
#8	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	97.7%
#9	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	96.2%
#10	GLM-4.7 glm-4.7	Z.AI	open	95.9%
#11	Kimi K2.5 kimi-k2.5	Moonshot AI	—	95.9%
#12	Kimi K2.6 kimi-k2.6	Moonshot AI	open	95.9%
#13	Qwen3.6 Max qwen3.6-max	Alibaba Cloud	closed	95.9%
#14	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	95.6%
#15	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	95.6%
#16	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	95.3%
#17	MiniMax M2.5 minimax-m2.5	MiniMax	open	95.3%
#18	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	95.0%
#19	MiMo-V2-Flash mimo-v2-flash	Xiaomi	open	95.0%
#20	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	95.0%
#21	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	94.7%
#22	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	94.4%
#23	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	94.2%
#24	Mistral Medium 3.5 mistral-medium-3-5	Mistral AI	open	94.2%
#25	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	94.2%
#26	GPT-5.5 gpt-5.5	OpenAI	closed	93.9%
#27	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	93.9%
#28	Qwen3.5 122B qwen3.5-122b-a10b	Alibaba Cloud	open	93.6%
#29	Grok 4.20 grok-4.20	xAI	closed	93.0%
#30	Kimi K2 Thinking kimi-k2-thinking	Moonshot AI	open	93.0%
#31	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	93.0%
#32	Nova 2 Pro nova-2-pro-preview	Amazon	closed	92.7%
#33	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	92.1%
#34	GPT-5.2 Codex gpt-5.2-codex	OpenAI	closed	92.1%
#35	Muse Spark muse-spark	Meta	closed	91.5%
#36	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	91.2%
#37	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	90.6%
#38	MiMo-V2.5 mimo-v2.5	Xiaomi	open	90.6%
#39	Grok 3 mini grok-3-mini	xAI	closed	90.4%
#40	Kimi K2.7 Code kimi-k2.7-code	Moonshot AI	open	90.1%
#41	Ling 2.6 1T ling-2.6-1t	InclusionAI	open	89.8%
#42	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	89.5%
#43	Qwen3.5 35B qwen3.5-35b-a3b	Alibaba Cloud	open	89.2%
#44	MiniMax M3 MiniMax-M3	MiniMax	—	88.9%
#45	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	88.6%
#46	Qwen3.5 Omni Plus qwen3.5-omni-plus	Alibaba Cloud	closed	88.3%
#47	Gemini 3 Pro gemini-3-pro	Google	closed	87.1%
#48	GPT-5.4 gpt-5.4	OpenAI	closed	87.1%
#49	GPT-5 Codex gpt-5-codex	OpenAI	closed	86.8%
#50	MiniMax M2 minimax-m2	MiniMax	open	86.8%
#51	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	86.3%
#52	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	86.0%
#53	MiniMax M2.1 minimax-m2.1	MiniMax	open	85.4%
#54	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	85.1%
#55	GPT-5 gpt-5	OpenAI	closed	84.8%
#56	GPT-5.2 gpt-5.2	OpenAI	closed	84.8%
#57	MiniMax M2.7 minimax-m2-7	MiniMax	open	84.8%
#58	Qwen3 Max qwen3-max	Alibaba Cloud	closed	83.6%
#59	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	83.3%
#60	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	83.3%
#61	GPT-5.1 gpt.5.1	OpenAI	closed	81.9%
#62	Command A+ command-a-plus	Cohere	open	80.7%
#63	o3 o3	OpenAI	closed	80.7%
#64	Gemini 3 Flash gemini-3-flash	Google	closed	80.4%
#65	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	78.1%
#66	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	76.0%
#67	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	75.7%
#68	Grok Code Fast 1 grok-code-fast-1	xAI	closed	75.7%
#69	Grok 4 grok-4	xAI	closed	74.9%
#70	Claude Opus 4 claude-opus-4	Anthropic	closed	73.4%
#71	Kimi K2 0905 kimi-k2-0905	Moonshot AI	open	73.4%
#72	Claude Opus 4.1 claude-opus-4-1	Anthropic	closed	71.4%
#73	GPT-5 mini gpt-5-mini	OpenAI	closed	71.1%
#74	GLM-4.6 glm-4.6	Z.AI	open	70.5%
#75	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	67.8%
#76	Grok 4 Fast grok-4-fast	xAI	closed	65.8%
#77	Claude Sonnet 4 claude-sonnet-4	Anthropic	closed	64.6%
#78	o1 o1	OpenAI	closed	62.6%
#79	Gemma 4 31B gemma-4-31b	Google	open	59.9%
#80	o4-mini o4-mini	OpenAI	closed	55.6%
#81	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	54.7%
#82	Claude Sonnet 3.7 claude-3-7-sonnet	Anthropic	closed	54.7%
#83	Gemini 2.5 Pro gemini-2.5-pro	Google	closed	54.1%
#84	Gemini 2.5 Flash gemini-2.5-flash	Google	closed	45.6%
#85	Gemma 4 26B gemma-4-26b-a4b	Google	open	43.6%
#86	Mistral Small 4 mistral-small-2603	Mistral AI	open	41.2%
#87	DeepSeek V3.1 Terminus deepseek-v3.1-terminus	DeepSeek	open	37.1%
#88	Gemma 4 12B gemma-4-12b	Google	open	36.3%
#89	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	31.3%
#90	Mistral Large 3 mistral-large-2512	Mistral AI	open	24.6%

The top of this snapshot is led by GLM-5.2 at 99.1%; third place is 0.58 points behind. The top-10 spread is 3.2 points.

Compare top models

GLM-5.2 vs Claude Fable 5 Claude Fable 5 vs GLM-5-Turbo GLM-5-Turbo vs GLM-5V-Turbo GLM-5V-Turbo vs GLM-5

About τ²-Bench Telecom

Definition and scoring

Organisation: Sierra / Artificial Analysis
Category: Agents
Version: 2
Direction: higher is better
Ranking use: Weighted 1.4%
Contamination risk: Low

Dual-control conversational agent benchmark for telecom support where agent and user must coordinate tool actions. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does τ²-Bench Telecom measure?

Dual-control conversational agent benchmark for telecom support where agent and user must coordinate tool actions.

Which model scores highest on τ²-Bench Telecom?

GLM-5.2 by Z.AI currently leads with 99.1%.

How many models are evaluated on τ²-Bench Telecom?

90 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.4% of its capability category. GLM-5.2 is overall #11.

Terminal-Bench

2.1 · 80 results · Weighted 1.6%

Berkeley Function-Calling Leaderboard

V3 · 2 results · Not Weighted

τ-bench

Rolling · 37 results · Weighted 1.6%

GAIA

Rolling · 13 results · Weighted 1.6%

BrowseComp

Rolling · 24 results · Weighted 1.6%

OSWorld

Rolling · 11 results · Weighted 1.6%

Agents · Benchmark profile

τ²-Bench Telecom

Dual-control conversational agent benchmark for telecom support where agent and user must coordinate tool actions.

Ranking weighted2Active90 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on τ²-Bench Telecom

Leaderboard · 90 models

GLM-5.2 leads at 99.1%, followed by Claude Fable 5 (98.5%) and GLM-5-Turbo (98.5%).

Ranking · 1.4%Agents2

#1Open

GLM-5.2

Z.AI

glm-5-2

99.1%

Overall #11 · Context 1M tokens

#2Closed

Claude Fable 5

Anthropic

claude-fable-5

98.5%

Overall #2 · Context 1M tokens

#3Closed

GLM-5-Turbo

Z.AI

glm-5-turbo

98.5%

Overall #28 · Context 200K tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

Z.AIAnthropicxAIAlibaba CloudDeepSeekMoonshot AIGoogleMiniMaxXiaomi

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
GLM-5.2	#1	Z.AI	99.1%
Claude Fable 5	#2	Anthropic	98.5%
GLM-5-Turbo	#3	Z.AI	98.5%
GLM-5V-Turbo	#4	Z.AI	98.5%
GLM-5	#5	Z.AI	98.2%
GLM-5.1	#6	Z.AI	97.7%
Grok 4.3	#7	xAI	97.7%
Qwen3.6 Plus	#8	Alibaba Cloud	97.7%
DeepSeek V4 Pro	#9	DeepSeek	96.2%
GLM-4.7	#10	Z.AI	95.9%
Kimi K2.5	#11	Moonshot AI	95.9%
Kimi K2.6	#12	Moonshot AI	95.9%
Qwen3.6 Max	#13	Alibaba Cloud	95.9%
Gemini 3.1 Pro Preview	#14	Google	95.6%
Qwen3.5 397B	#15	Alibaba Cloud	95.6%
Gemini 3.5 Flash	#16	Google	95.3%
MiniMax M2.5	#17	MiniMax	95.3%
DeepSeek V4 Flash	#18	DeepSeek	95.0%
MiMo-V2-Flash	#19	Xiaomi	95.0%
MiMo-V2-Pro	#20	Xiaomi	95.0%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
GLM-5.2	#1	Z.AI	99.1%
Claude Fable 5	#2	Anthropic	98.5%
GLM-5-Turbo	#3	Z.AI	98.5%
GLM-5V-Turbo	#4	Z.AI	98.5%
GLM-5	#5	Z.AI	98.2%
GLM-5.1	#6	Z.AI	97.7%
Grok 4.3	#7	xAI	97.7%
Qwen3.6 Plus	#8	Alibaba Cloud	97.7%
DeepSeek V4 Pro	#9	DeepSeek	96.2%
GLM-4.7	#10	Z.AI	95.9%
Kimi K2.5	#11	Moonshot AI	95.9%
Kimi K2.6	#12	Moonshot AI	95.9%
Qwen3.6 Max	#13	Alibaba Cloud	95.9%
Gemini 3.1 Pro Preview	#14	Google	95.6%
Qwen3.5 397B	#15	Alibaba Cloud	95.6%
Gemini 3.5 Flash	#16	Google	95.3%
MiniMax M2.5	#17	MiniMax	95.3%
DeepSeek V4 Flash	#18	DeepSeek	95.0%
MiMo-V2-Flash	#19	Xiaomi	95.0%
MiMo-V2-Pro	#20	Xiaomi	95.0%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Anthropic	1	98.5%
Z.AI	6	98.0%
xAI	1	97.7%
Alibaba Cloud	3	96.4%
Moonshot AI	2	95.9%
DeepSeek	2	95.6%
Google	2	95.5%
MiniMax	1	95.3%
Xiaomi	2	95.0%

Benchmark score table (90 models)

One best score per model · higher is better

τ²-Bench Telecom leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	GLM-5.2 glm-5-2	Z.AI	open	99.1%
#2	Claude Fable 5 claude-fable-5	Anthropic	closed	98.5%
#3	GLM-5-Turbo glm-5-turbo	Z.AI	closed	98.5%
#4	GLM-5V-Turbo glm-5v-turbo	Z.AI	closed	98.5%
#5	GLM-5 glm-5	Z.AI	open	98.2%
#6	GLM-5.1 glm-5-1	Z.AI	open	97.7%
#7	Grok 4.3 grok-4.3	xAI	closed	97.7%
#8	Qwen3.6 Plus qwen3-6-plus	Alibaba Cloud	closed	97.7%
#9	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	96.2%
#10	GLM-4.7 glm-4.7	Z.AI	open	95.9%
#11	Kimi K2.5 kimi-k2.5	Moonshot AI	—	95.9%
#12	Kimi K2.6 kimi-k2.6	Moonshot AI	open	95.9%
#13	Qwen3.6 Max qwen3.6-max	Alibaba Cloud	closed	95.9%
#14	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	95.6%
#15	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	95.6%
#16	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	95.3%
#17	MiniMax M2.5 minimax-m2.5	MiniMax	open	95.3%
#18	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	95.0%
#19	MiMo-V2-Flash mimo-v2-flash	Xiaomi	open	95.0%
#20	MiMo-V2-Pro mimo-v2-pro	Xiaomi	closed	95.0%
#21	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	—	94.7%
#22	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	94.4%
#23	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	94.2%
#24	Mistral Medium 3.5 mistral-medium-3-5	Mistral AI	open	94.2%
#25	Qwen3.6 27B qwen3.6-27b	Alibaba Cloud	open	94.2%
#26	GPT-5.5 gpt-5.5	OpenAI	closed	93.9%
#27	Qwen3.5 27B qwen3.5-27b	Alibaba Cloud	open	93.9%
#28	Qwen3.5 122B qwen3.5-122b-a10b	Alibaba Cloud	open	93.6%
#29	Grok 4.20 grok-4.20	xAI	closed	93.0%
#30	Kimi K2 Thinking kimi-k2-thinking	Moonshot AI	open	93.0%
#31	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	93.0%
#32	Nova 2 Pro nova-2-pro-preview	Amazon	closed	92.7%
#33	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	92.1%
#34	GPT-5.2 Codex gpt-5.2-codex	OpenAI	closed	92.1%
#35	Muse Spark muse-spark	Meta	closed	91.5%
#36	MiMo-V2-Omni mimo-v2-omni	Xiaomi	closed	91.2%
#37	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	90.6%
#38	MiMo-V2.5 mimo-v2.5	Xiaomi	open	90.6%
#39	Grok 3 mini grok-3-mini	xAI	closed	90.4%
#40	Kimi K2.7 Code kimi-k2.7-code	Moonshot AI	open	90.1%
#41	Ling 2.6 1T ling-2.6-1t	InclusionAI	open	89.8%
#42	Claude Opus 4.5 claude-opus-4-5	Anthropic	closed	89.5%
#43	Qwen3.5 35B qwen3.5-35b-a3b	Alibaba Cloud	open	89.2%
#44	MiniMax M3 MiniMax-M3	MiniMax	—	88.9%
#45	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	88.6%
#46	Qwen3.5 Omni Plus qwen3.5-omni-plus	Alibaba Cloud	closed	88.3%
#47	Gemini 3 Pro gemini-3-pro	Google	closed	87.1%
#48	GPT-5.4 gpt-5.4	OpenAI	closed	87.1%
#49	GPT-5 Codex gpt-5-codex	OpenAI	closed	86.8%
#50	MiniMax M2 minimax-m2	MiniMax	open	86.8%
#51	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	86.3%
#52	GPT-5.3-Codex gpt-5.3-codex	OpenAI	closed	86.0%
#53	MiniMax M2.1 minimax-m2.1	MiniMax	open	85.4%
#54	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	85.1%
#55	GPT-5 gpt-5	OpenAI	closed	84.8%
#56	GPT-5.2 gpt-5.2	OpenAI	closed	84.8%
#57	MiniMax M2.7 minimax-m2-7	MiniMax	open	84.8%
#58	Qwen3 Max qwen3-max	Alibaba Cloud	closed	83.6%
#59	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	83.3%
#60	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	83.3%
#61	GPT-5.1 gpt.5.1	OpenAI	closed	81.9%
#62	Command A+ command-a-plus	Cohere	open	80.7%
#63	o3 o3	OpenAI	closed	80.7%
#64	Gemini 3 Flash gemini-3-flash	Google	closed	80.4%
#65	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	78.1%
#66	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	76.0%
#67	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	75.7%
#68	Grok Code Fast 1 grok-code-fast-1	xAI	closed	75.7%
#69	Grok 4 grok-4	xAI	closed	74.9%
#70	Claude Opus 4 claude-opus-4	Anthropic	closed	73.4%
#71	Kimi K2 0905 kimi-k2-0905	Moonshot AI	open	73.4%
#72	Claude Opus 4.1 claude-opus-4-1	Anthropic	closed	71.4%
#73	GPT-5 mini gpt-5-mini	OpenAI	closed	71.1%
#74	GLM-4.6 glm-4.6	Z.AI	open	70.5%
#75	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	67.8%
#76	Grok 4 Fast grok-4-fast	xAI	closed	65.8%
#77	Claude Sonnet 4 claude-sonnet-4	Anthropic	closed	64.6%
#78	o1 o1	OpenAI	closed	62.6%
#79	Gemma 4 31B gemma-4-31b	Google	open	59.9%
#80	o4-mini o4-mini	OpenAI	closed	55.6%
#81	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	54.7%
#82	Claude Sonnet 3.7 claude-3-7-sonnet	Anthropic	closed	54.7%
#83	Gemini 2.5 Pro gemini-2.5-pro	Google	closed	54.1%
#84	Gemini 2.5 Flash gemini-2.5-flash	Google	closed	45.6%
#85	Gemma 4 26B gemma-4-26b-a4b	Google	open	43.6%
#86	Mistral Small 4 mistral-small-2603	Mistral AI	open	41.2%
#87	DeepSeek V3.1 Terminus deepseek-v3.1-terminus	DeepSeek	open	37.1%
#88	Gemma 4 12B gemma-4-12b	Google	open	36.3%
#89	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	31.3%
#90	Mistral Large 3 mistral-large-2512	Mistral AI	open	24.6%

The top of this snapshot is led by GLM-5.2 at 99.1%; third place is 0.58 points behind. The top-10 spread is 3.2 points.

Compare top models

GLM-5.2 vs Claude Fable 5 Claude Fable 5 vs GLM-5-Turbo GLM-5-Turbo vs GLM-5V-Turbo GLM-5V-Turbo vs GLM-5

About τ²-Bench Telecom

Definition and scoring

Organisation: Sierra / Artificial Analysis
Category: Agents
Version: 2
Direction: higher is better
Ranking use: Weighted 1.4%
Contamination risk: Low

Open benchmark source ↗

FAQ

Quick answers

What does τ²-Bench Telecom measure?

Dual-control conversational agent benchmark for telecom support where agent and user must coordinate tool actions.

Which model scores highest on τ²-Bench Telecom?

GLM-5.2 by Z.AI currently leads with 99.1%.

How many models are evaluated on τ²-Bench Telecom?

90 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.4% of its capability category. GLM-5.2 is overall #11.

Terminal-Bench

2.1 · 80 results · Weighted 1.6%

Berkeley Function-Calling Leaderboard

V3 · 2 results · Not Weighted

τ-bench

Rolling · 37 results · Weighted 1.6%

GAIA

Rolling · 13 results · Weighted 1.6%

BrowseComp

Rolling · 24 results · Weighted 1.6%

OSWorld

Rolling · 11 results · Weighted 1.6%

Leaderboard · 90 models

GLM-5.2

Claude Fable 5

GLM-5-Turbo

See the field, not just the table.

Benchmark score table (90 models)

Compare top models

Definition and scoring

Quick answers

What does τ²-Bench Telecom measure?

Which model scores highest on τ²-Bench Telecom?

How many models are evaluated on τ²-Bench Telecom?

Does this affect overall Lumina rank?

More Agents benchmarks

Leaderboard · 90 models

GLM-5.2

Claude Fable 5

GLM-5-Turbo

See the field, not just the table.

Benchmark score table (90 models)

Compare top models

Definition and scoring

Quick answers

What does τ²-Bench Telecom measure?

Which model scores highest on τ²-Bench Telecom?

How many models are evaluated on τ²-Bench Telecom?

Does this affect overall Lumina rank?

More Agents benchmarks