Agent ranking

Best AI models for agents

Language models measured on at least two public agent boards (LMArena Agent, BFCL V4, WildClawBench), ordered by the agent component of the ModelCap Index. Models with a single agent result follow in their own tier, ordered by that board, so one result never outranks two.

Snapshot as of 24 October 2026

Claude Opus 5.5 from Anthropic leads the best-for-agents list as of 24 October 2026 with LMArena Agent 20.8% on one public agent board, holding ModelCap Index position #1; 31 models qualify.

Models on this list

With openly downloadable weights

With a listed API price

Best AI models for agents (0)

Every row is a current, canonical language model with a public position on the ModelCap Index; the list filters by a published fact and orders by a published figure.

No ranked model currently holds admitted results on two or more public agent boards, so the primary tier is empty at this refresh. The single-board tier below is the complete agent evidence the Index admits right now.

Measured on one agent board (31)

Ranked models with a single admitted agent result, ordered by that board's normalised score. One board is a narrower claim than two, so these rows sit apart from the list above and never mix into it.

Measured on one agent board: Ranked models with a single admitted agent result, ordered by that board's normalised score. One board is a narrower claim than two, so these rows sit apart from the list above and never mix into it.
#	Model	Agent boards	ModelCap Index	Input $/1M	Output $/1M	Context	Weights
1	Claude Opus 5.5Anthropic	LMArena Agent 20.8%	#187.3	$4.00	$20.00	1M	No linked weights
2	Claude Fable 5.1Anthropic	LMArena Agent 13.9%	#385.6	$10.00	$50.00	1M	No linked weights
3	GPT-6 AstraOpenAI	LMArena Agent 14.0%	#1379.9	$10.00	$50.00	1M	No linked weights
4	Claude Sonnet 5.5Anthropic	LMArena Agent 14.0%	#1879.3	$2.00	$10.00	1M	No linked weights
5	GPT-6.1 SolOpenAI	LMArena Agent 13.6%	#880.7	$2.00	$10.00	1M	No linked weights
6	Gemini 4 ArgonGoogle	LMArena Agent 11.7%	#287.2	$2.00	$10.00	1M	No linked weights
7	MiMo-V2.6-ProXiaomi	LMArena Agent 3.3%	#2079.0	$0.435	$0.87	1M	Open weights
8	GPT-6 LunaOpenAI	LMArena Agent 2.8%	#4966.8	$0.10	$0.50	1M	No linked weights
9	Muse Spark 1.3Meta	LMArena Agent 3.1%	#583.7	$1.25	$4.25	1M	No linked weights
10	Gemini 3.8 FlashGoogle	LMArena Agent 3.2%	#682.6	$0.825	$4.13	1M	No linked weights
11	Muse WonderMeta	LMArena Agent 3.3%	#485.4	$4.00	$20.00	1M	No linked weights
12	Grok 4.7xAI	LMArena Agent 3.0%	#5066.0	$2.00	$6.00	500K	No linked weights
13	MiMo-V2.6-FlashXiaomi	LMArena Agent 2.1%	#4269.5	$0.14	$0.28	1M	Open weights
14	Hy4 previewTencent	LMArena Agent 2.0%	#4070.8	$0.751	$2.25	1M	Open weights
15	Kimi K3Moonshot AI	LMArena Agent 1.6%	#1080.4	$0.239	$10.80	1M	Restricted license
16	Claude Haiku 5.5Anthropic	LMArena Agent 1.1%	#7460.1	$0.10	$0.50	1M	No linked weights
17	GLM 5.3Z.ai	LMArena Agent 1.2%	#1979.0	$0.091	$1.61	1M	Restricted license
18	GPT-5.6 TerraOpenAI	LMArena Agent 0.2% · AA τ²-Bench Telecom 86.3%	#3075.4	$2.00	$12.00	1M	No linked weights
19	GLM 5.3 FlashZ.ai	LMArena Agent -1.8%	#2477.3	$0.15	$0.50	1M	Open weights
20	Step 5 PreviewStepFun	LMArena Agent -2.0%	#4468.4	$1.00	$2.70	1M	Restricted license
21	Beam 1.0 501B A23BReflection	LMArena Agent -3.0%	#5664.4	$0.25	$1.00	1M	Open weights
22	Qwen3.8 27BQwen	LMArena Agent -4.1%	#6163.3	$0.20	$2.50	1M	Open weights
23	Gemini 3.1 Pro PreviewGoogle	LMArena Agent -4.4% · AA τ²-Bench Telecom 95.6%	#1479.9	$2.00	$12.00	1M	No linked weights
24	Mistral Large 4Mistral AI	LMArena Agent -5.1%	#6263.2	$1.36	$4.18	1M	No linked weights
25	Qwen3.7 PlusQwen	LMArena Agent -7.4% · AA τ²-Bench Telecom 93.0%	#3671.2	$0.32	$1.28	1M	No linked weights
26	MiniMax M3MiniMax	LMArena Agent -10.1% · AA τ²-Bench Telecom 88.9%	#5165.7	$0.30	$1.20	1M	Restricted license
27	Mistral Medium 3.5Mistral AI	LMArena Agent -12.1% · AA τ²-Bench Telecom 94.2%	#7759.6	$1.50	$7.50	262K	No linked weights
28	Nemotron 3 UltraNVIDIA	LMArena Agent -14.7%	#24627.3	$0.60	$2.40	262K	Restricted license
29	InklingThinking Machines	LMArena Agent -14.9%	#5365.2	$1.00	$4.05	524K	Open weights
30	Inkling SmallThinking Machines	LMArena Agent -16.1%	#10952.0	$0.45	$1.20	524K	Open weights
31	Solar Pro 4Upstage	LMArena Agent -20.2%	#12448.0	$0.15	$0.60	524K	No linked weights

Best AI models for agents: common questions

Which model tops the best-for-agents list right now?

Claude Opus 5.5 (Anthropic), as of 24 October 2026, with LMArena Agent 20.8% on one public agent board and ModelCap Index position #1.

How is the best-for-agents list built?

It contains current, canonical language models with a measured public ModelCap Index position and admitted results on at least two public agent-family boards, ordered by the agent component of the ModelCap Index (ties broken by Index rank). A second tier, "Measured on one agent board", follows: Ranked models with a single admitted agent result, ordered by that board's normalised score. One board is a narrower claim than two, so these rows sit apart from the list above and never mix into it. Nothing is estimated for the list itself: the filter is a published fact and the order is a published figure, and every row links to the model page that shows its evidence.

How many models qualify for the best-for-agents list?

31 models qualify (0 in the primary tier, 31 measured on one board). The population is the live ModelCap Index, so a model appears here the moment it holds a public position and meets the filter.

What is the cheapest model on the best-for-agents list?

MiMo-V2.6-Flash lists the lowest blended API price at $0.175 per 1M tokens ($0.14 input / $0.28 output).

Which is the best open-weight model on this list?

MiMo-V2.6-Pro from Xiaomi is the highest-placed model with openly downloadable weights (#7 here, ModelCap Index #20).

How current is the best-for-agents list?

Positions, prices and context figures come from the same published snapshot dated 24 October 2026. Publication and page updates happen separately; use the snapshot date shown on the page.