Best LLM for agents

Agentic tool use covers whether a model drives tools to a finished outcome, stays steerable, and recovers when a command fails. Terminal-Bench and other evaluator-run agentic results identify the model together with its agent harness and effort; they are not pure model-only scores.

Aldena runs these models inside your team rooms. See what each one costs.

available on aldena

122 of 122 ranked models

Ranked models
rank	model	vendor	composite	pricein / out	benchmarks	%	%	%	score	score	score	score	score
1	Claude Fable 5 (high)anthropic/claude-fable-5:high	Anthropic	81.0	$10.00 / $50.00	6 of 8 benchmarks		80.4%		12.0	8.8	10.7	1.2	14.2
2	Claude Opus 5 (high)anthropic/claude-opus-5:high	Anthropic	78.0	$5.00 / $25.00	5 of 8 benchmarks				12.2	11.3	15.3	1.1	13.9
3	GPT-5.6 Sol (xhigh)openai/gpt-5.6-sol:xhigh	OpenAI	77.1	$5.00 / $30.00	5 of 8 benchmarks				10.9	9.2	10.1	1.2	10.7
4	Claude Opus 5 (max)anthropic/claude-opus-5:max	Anthropic	75.9	$5.00 / $25.00	5 of 8 benchmarks				11.9	7.0	18.0	1.2	14.4
5	Kimi K3 (max)moonshotai/kimi-k3:max	MoonshotAI	75.7	$3.00 / $15.00	6 of 8 benchmarks			82.3%	10.6	7.8	15.6	1.2	8.1
6	Claude Opus 4.6anthropic/claude-opus-4.6	Anthropic	74.0	$5.00 / $25.00	6 of 8 benchmarks	75.6%			6.9	8.3	5.3	1.2	11.5
7	GPT-5.5 (xhigh)openai/gpt-5.5:xhigh	OpenAI	73.3	$5.00 / $30.00	7 of 8 benchmarks		83.1%	75.3%	8.9	9.2	5.0	1.2	14.5
8	GPT-5.5 (high)openai/gpt-5.5:high	OpenAI	73.3	$5.00 / $30.00	5 of 8 benchmarks				7.7	8.6	4.0	1.2	13.2
9	Grok 4.5x-ai/grok-4.5	SpaceXAI	69.4	$2.00 / $6.00	5 of 8 benchmarks				6.2	7.0	6.4	1.2	11.2
10	Claude Opus 4.7 (high)anthropic/claude-opus-4.7:high	Anthropic	69.1	$5.00 / $25.00	5 of 8 benchmarks				8.2	7.7	6.6	1.1	13.1
11	Claude Opus 4.7anthropic/claude-opus-4.7	Anthropic	68.8	$5.00 / $25.00	5 of 8 benchmarks				7.7	9.9	5.5	1.2	10.2
12	GPT-5.5openai/gpt-5.5	OpenAI	68.8	$5.00 / $30.00	5 of 8 benchmarks				6.3	7.4	3.6	1.2	11.9
13	Claude Fable 5 (xhigh)anthropic/claude-fable-5:xhigh	Anthropic	66.8	$10.00 / $50.00	1 of 8 benchmarks		83.8%
14	GLM 5.2 (max)z-ai/glm-5.2:max	Z.ai	66.1	$0.462 / $1.452	5 of 8 benchmarks				6.7	6.1	8.4	1.2	6.3
15	Gemini 3 Flash Preview (high)google/gemini-3-flash-preview:high	Google	65.7	$0.50 / $3.00	1 of 8 benchmarks	75.8%
16	MiniMax M2.5 (high)minimax/minimax-m2.5:high	MiniMax	65.7	$0.22 / $0.90	1 of 8 benchmarks	75.8%
17	Claude Opus 5 (xhigh)anthropic/claude-opus-5:xhigh	Anthropic	65.6	$5.00 / $25.00	1 of 8 benchmarks			85.8%
18	GPT-5.4 (high)openai/gpt-5.4:high	OpenAI	64.3	$2.50 / $15.00	5 of 8 benchmarks				5.0	6.0	4.3	1.2	9.6
19	Claude Opus 4.5 (medium)anthropic/claude-opus-4.5:medium	Anthropic	63.8	$5.00 / $25.00	1 of 8 benchmarks	74.4%
20	Claude Fable 5anthropic/claude-fable-5	Anthropic	63.3	$10.00 / $50.00	1 of 8 benchmarks			83.3%
21	Claude Opus 4.8 (high)anthropic/claude-opus-4.8:high	Anthropic	63.1	$5.00 / $25.00	6 of 8 benchmarks		78.9%		9.8	8.4	9.4	-0.8	9.3
22	GPT-5.2-Codexopenai/gpt-5.2-codex	OpenAI	61.6	$1.75 / $14.00	1 of 8 benchmarks	72.8%
23	GPT-5.2 (high)openai/gpt-5.2:high	OpenAI	61.6	$1.75 / $14.00	1 of 8 benchmarks	72.8%
24	GLM 5 (high)z-ai/glm-5:high	Z.ai	61.6	$0.60 / $1.92	1 of 8 benchmarks	72.8%
25	GPT-5.6 Luna (xhigh)openai/gpt-5.6-luna:xhigh	OpenAI	60.8	$0.10 / $0.60	5 of 8 benchmarks				4.3	1.5	-1.1	1.2	11.7
26	Claude Opus 4.8 (max)anthropic/claude-opus-4.8:max	Anthropic	60.5	$5.00 / $25.00	1 of 8 benchmarks			82.2%
27	Muse Sparkmeta/muse-spark	Meta	60.5	—	1 of 8 benchmarks			82.2%
28	DeepSeek V4 Flash 0423 (high)deepseek/deepseek-v4-flash:high	DeepSeek	60.3	$0.0643 / $0.1285	5 of 8 benchmarks				4.0	3.3	8.7	1.2	4.5
29	Muse Spark 1.1meta/muse-spark-1.1	Meta	60.2	$1.25 / $4.25	6 of 8 benchmarks			88.1%	1.2	-3.3	7.3	1.2	5.6
30	GPT-5.6 Terra (xhigh)openai/gpt-5.6-terra:xhigh	OpenAI	60.2	$1.00 / $6.00	5 of 8 benchmarks				3.8	6.2	-1.8	1.2	9.7
31	Qwen3.8 Maxqwen/qwen3.8-max	Qwen	60.2	$2.00 / $6.00	5 of 8 benchmarks				7.6	5.3	12.5	0.1	8.4
32	Claude Sonnet 4.5 (high)anthropic/claude-sonnet-4.5:high	Anthropic	60.1	$3.00 / $15.00	1 of 8 benchmarks	71.4%
33	Claude Opus 4.5 (high)anthropic/claude-opus-4.5:high	Anthropic	59.6	$5.00 / $25.00	2 of 8 benchmarks	76.8%		69.8%
34	Kimi K2.5 (high)moonshotai/kimi-k2.5:high	MoonshotAI	59.4	$0.57 / $2.85	1 of 8 benchmarks	70.8%
35	GPT-5.6 Solopenai/gpt-5.6-sol	OpenAI	58.7	$5.00 / $30.00	1 of 8 benchmarks			81.8%
36	Claude Sonnet 4.5anthropic/claude-sonnet-4.5	Anthropic	58.6	$3.00 / $15.00	1 of 8 benchmarks	70.6%
37	Grok 4.5 (high)x-ai/grok-4.5:high	SpaceXAI	58.4	$2.00 / $6.00	1 of 8 benchmarks		79.3%
38	DeepSeek V3.2 (high)deepseek/deepseek-v3.2:high	DeepSeek	57.9	$0.269 / $0.40	1 of 8 benchmarks	70.0%
39	Gemini 3.7 Flash (high)google/gemini-3.7-flash:high	Google	57.8	$0.375 / $1.875	5 of 8 benchmarks				3.6	2.8	9.8	1.2	2.3
40	Gemini 3 Pro Previewgoogle/gemini-3-pro-preview	Google	57.6	—	2 of 8 benchmarks	74.2%		70.3%
41	Inkling Smallthinkingmachines/inkling-small	Thinking Machines	57.6	$0.45 / $1.20	1 of 8 benchmarks			79.2%
42	Claude Opus 4.8anthropic/claude-opus-4.8	Anthropic	57.2	$5.00 / $25.00	5 of 8 benchmarks				2.1	8.3	8.8	-31.0	10.9
43	Gemini 3 Pro Preview (high)google/gemini-3-pro-preview:high	Google	57.1	—	1 of 8 benchmarks	69.6%
44	Claude Sonnet 5 (high)anthropic/claude-sonnet-5:high	Anthropic	57.1	$2.00 / $10.00	6 of 8 benchmarks		74.6%		7.1	5.2	3.1	1.1	11.2
45	GPT-5.2openai/gpt-5.2	OpenAI	56.4	$1.75 / $14.00	1 of 8 benchmarks	69.0%
46	Claude Opus 4anthropic/claude-opus-4	Anthropic	55.7	$15.00 / $75.00	1 of 8 benchmarks	67.6%
47	Claude Haiku 4.5 (high)anthropic/claude-haiku-4.5:high	Anthropic	54.9	$1.00 / $5.00	1 of 8 benchmarks	66.6%
48	GLM 5.2z-ai/glm-5.2	Z.ai	54.1	$0.462 / $1.452	1 of 8 benchmarks			77.8%
49	GPT-5.1-Codex (medium)openai/gpt-5.1-codex:medium	OpenAI	53.8	$1.25 / $10.00	1 of 8 benchmarks	66.0%
50	GPT-5.1 (medium)openai/gpt-5.1:medium	OpenAI	53.8	$1.25 / $10.00	1 of 8 benchmarks	66.0%
51	Claude Opus 4.6 (max)anthropic/claude-opus-4.6:max	Anthropic	53.0	$5.00 / $25.00	1 of 8 benchmarks			76.8%
52	GPT-5.6 Terra (max)openai/gpt-5.6-terra:max	OpenAI	52.9	$1.00 / $6.00	1 of 8 benchmarks		78.4%
53	Kimi K2.7 Codemoonshotai/kimi-k2.7-code	MoonshotAI	52.7	$0.71 / $3.50	5 of 8 benchmarks				1.1	-1.8	4.4	1.2	-1.2
54	GPT-5 (medium)openai/gpt-5:medium	OpenAI	52.7	$1.25 / $10.00	1 of 8 benchmarks	65.0%
55	Claude Sonnet 4.6anthropic/claude-sonnet-4.6	Anthropic	52.1	$3.00 / $15.00	6 of 8 benchmarks			69.5%	3.1	2.2	0.1	1.1	11.2
56	Claude Sonnet 4anthropic/claude-sonnet-4	Anthropic	52.0	$3.00 / $15.00	1 of 8 benchmarks	64.9%
57	Inkling (xhigh)thinkingmachines/inkling:xhigh	Thinking Machines	51.8	$0.95 / $4.05	1 of 8 benchmarks			76.0%
58	Kimi K2 Thinkingmoonshotai/kimi-k2-thinking	MoonshotAI	51.2	$0.60 / $2.50	1 of 8 benchmarks	63.4%
59	MiniMax M2minimax/minimax-m2	MiniMax	50.5	$0.255 / $1.02	1 of 8 benchmarks	61.0%
60	Muse Spark 1.1 (xhigh)meta/muse-spark-1.1:xhigh	Meta	50.1	$1.25 / $4.25	1 of 8 benchmarks		76.2%
61	DeepSeek V3.2 (thinking)deepseek/deepseek-v3.2:thinking	DeepSeek	49.7	$0.269 / $0.40	1 of 8 benchmarks	60.0%
62	GPT-5 Mini (medium)openai/gpt-5-mini:medium	OpenAI	49.0	$0.25 / $2.00	1 of 8 benchmarks	59.8%
63	GPT-5.4 (xhigh)openai/gpt-5.4:xhigh	OpenAI	48.4	$2.50 / $15.00	1 of 8 benchmarks			70.6%
64	o3openai/o3	OpenAI	48.3	$2.00 / $8.00	1 of 8 benchmarks	58.4%
65	Kimi K2.6moonshotai/kimi-k2.6	MoonshotAI	47.7	$0.5415 / $2.28	5 of 8 benchmarks				-0.6	-1.1	0.6	1.2	-5.5
66	Gemini 3.5 Flash (high)google/gemini-3.5-flash:high	Google	47.6	$1.50 / $9.00	6 of 8 benchmarks			83.6%	-0.3	-0.5	0.6	0.3	-1.7
67	Devstral Small 2512mistralai/devstral-small-2512	Mistral AI	47.5	—	1 of 8 benchmarks	56.4%
68	GPT-5.6 Luna (max)openai/gpt-5.6-luna:max	OpenAI	47.3	$0.10 / $0.60	1 of 8 benchmarks		75.7%
69	GPT-5 Miniopenai/gpt-5-mini	OpenAI	46.8	$0.25 / $2.00	1 of 8 benchmarks	56.2%
70	Claude Opus 4.7 (max)anthropic/claude-opus-4.7:max	Anthropic	46.5	$5.00 / $25.00	2 of 8 benchmarks		68.9%	79.1%
71	Qwen3 Coder 480B A35b Instructqwen/qwen3-coder-480b-a35b-instruct	Qwen	45.7	—	1 of 8 benchmarks	55.4%
72	GLM 4.6z-ai/glm-4.6	Z.ai	45.7	$0.55 / $2.20	1 of 8 benchmarks	55.4%
73	Qwen3.7 Maxqwen/qwen3.7-max	Qwen	45.6	$1.475 / $4.425	5 of 8 benchmarks				0.2	-0.0	-0.3	0.6	5.0
74	GLM 4.5z-ai/glm-4.5	Z.ai	44.5	$0.60 / $2.20	1 of 8 benchmarks	54.2%
75	Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview	Google	44.4	$2.00 / $12.00	5 of 8 benchmarks				-0.4	3.2	1.5	0.9	-11.5
76	GLM 5.1z-ai/glm-5.1	Z.ai	44.0	$0.966 / $3.036	6 of 8 benchmarks			75.6%	0.6	1.7	1.8	-0.5	-0.9
77	Devstral 2mistralai/devstral-2	Mistral AI	43.8	—	1 of 8 benchmarks	53.8%
78	GPT-5.2 (xhigh)openai/gpt-5.2:xhigh	OpenAI	43.8	$1.75 / $14.00	1 of 8 benchmarks			67.6%
79	Gemini 3.1 Pro Preview (high)google/gemini-3.1-pro-preview:high	Google	43.5	$2.00 / $12.00	2 of 8 benchmarks		65.8%	78.2%
80	Gemini 2.5 Progoogle/gemini-2.5-pro	Google	43.1	$1.25 / $10.00	1 of 8 benchmarks	53.6%
81	Kimi K2p5moonshotai/kimi-k2p5	Moonshot AI	42.6	—	1 of 8 benchmarks			64.4%
82	Claude 3.7 Sonnetanthropic/claude-3-7-sonnet	Anthropic	42.3	—	1 of 8 benchmarks	52.8%
83	Gemini 3 Pro (high)google/gemini-3-pro:high	Google	41.8	—	1 of 8 benchmarks		73.9%
84	DeepSeek V4 Prodeepseek/deepseek-v4-pro	DeepSeek	41.7	$1.168 / $2.336	5 of 8 benchmarks				0.1	0.6	-2.2	0.3	4.4
85	o4 Miniopenai/o4-mini	OpenAI	41.6	$1.10 / $4.40	1 of 8 benchmarks	45.0%
86	Kimi K2 Instructmoonshotai/kimi-k2-instruct	Moonshot AI	40.8	—	1 of 8 benchmarks	43.8%
87	Claude Sonnet 4.5 (thinking)anthropic/claude-sonnet-4.5:thinking	Anthropic	40.3	$3.00 / $15.00	1 of 8 benchmarks			59.5%
88	Gemini 3.6 Flash (high)google/gemini-3.6-flash:high	Google	40.2	$0.75 / $3.75	5 of 8 benchmarks				-2.4	-4.4	-0.9	1.2	-3.6
89	GPT-4.1openai/gpt-4.1	OpenAI	40.1	$2.00 / $8.00	1 of 8 benchmarks	39.6%
90	GPT-5 Nano (medium)openai/gpt-5-nano:medium	OpenAI	39.4	$0.05 / $0.40	1 of 8 benchmarks	34.8%
91	GLM 4.7z-ai/glm-4.7	Z.ai	39.2	$0.40 / $1.75	1 of 8 benchmarks			58.1%
92	Gemini 2.5 Flashgoogle/gemini-2.5-flash	Google	38.6	$0.30 / $2.50	1 of 8 benchmarks	28.7%
93	Qwen3.7 Plusqwen/qwen3.7-plus	Qwen	38.4	$0.32 / $1.28	5 of 8 benchmarks				-2.0	-5.6	-1.1	0.2	6.3
94	Gemini 3.1 Flash Lite (high)google/gemini-3.1-flash-lite:high	Google	38.0	$0.25 / $1.50	1 of 8 benchmarks			57.1%
95	gpt-oss-120bopenai/gpt-oss-120b	OpenAI	37.9	$0.03 / $0.17	1 of 8 benchmarks	26.0%
96	MiniMax M3minimax/minimax-m3	MiniMax	37.5	$0.30 / $1.20	5 of 8 benchmarks				-2.5	-5.1	-5.8	0.7	5.9
97	GPT-4.1 Miniopenai/gpt-4.1-mini	OpenAI	37.1	$0.40 / $1.60	1 of 8 benchmarks	23.9%
98	GPT-5.4 Mini (xhigh)openai/gpt-5.4-mini:xhigh	OpenAI	36.9	$0.75 / $4.50	1 of 8 benchmarks			56.7%
99	GPT-4o (2024-11-20)openai/gpt-4o-2024-11-20	OpenAI	36.4	$2.50 / $10.00	1 of 8 benchmarks	21.6%
100	GPT-5.1 (high)openai/gpt-5.1:high	OpenAI	35.7	$1.25 / $10.00	1 of 8 benchmarks			50.1%
101	Llama 4 Maverick 17B 128e Instructmeta-llama/llama-4-maverick-17b-128e-instruct	Meta	35.7	—	1 of 8 benchmarks	21.0%
102	DeepSeek V4 Flash 0423deepseek/deepseek-v4-flash	DeepSeek	35.5	$0.0643 / $0.1285	5 of 8 benchmarks				-2.2	-1.0	-2.0	-1.3	2.6
103	MiMo-V2.5-Proxiaomi/mimo-v2.5-pro	Xiaomi	35.5	$0.435 / $0.87	5 of 8 benchmarks				-2.0	-2.0	-2.8	0.0	1.6
104	Gemini 2.0 Flash 001google/gemini-2.0-flash-001	Google	34.9	—	1 of 8 benchmarks	13.5%
105	o3 Proopenai/o3-pro	OpenAI	34.6	$20.00 / $80.00	1 of 8 benchmarks			44.5%
106	Llama 4 Scout 17B 16e Instructmeta-llama/llama-4-scout-17b-16e-instruct	Meta	34.2	—	1 of 8 benchmarks	9.1%
107	Claude Haiku 4.5anthropic/claude-haiku-4.5	Anthropic	33.4	$1.00 / $5.00	1 of 8 benchmarks			40.2%
108	Qwen2.5 Coder 32B Instructqwen/qwen2.5-coder-32b-instruct	Qwen	33.4	—	1 of 8 benchmarks	9.0%
109	GLM 5.1 (max)z-ai/glm-5.1:max	Z.ai	33.4	$0.966 / $3.036	1 of 8 benchmarks		58.7%
110	Gemini 3.5 Flash (medium)google/gemini-3.5-flash:medium	Google	33.2	$1.50 / $9.00	5 of 8 benchmarks				-3.5	-3.5	-8.3	0.5	-0.3
111	Hy3tencent/hy3	Tencent	32.5	$0.132 / $0.528	5 of 8 benchmarks				-1.3	-8.6	-2.5	-1.3	4.0
112	Inklingthinkingmachines/inkling	Thinking Machines	32.2	$0.95 / $4.05	5 of 8 benchmarks				-6.6	-11.8	-11.7	0.6	6.8
113	Grok 4.3 (high)x-ai/grok-4.3:high	SpaceXAI	30.9	$1.25 / $2.50	5 of 8 benchmarks				-8.5	-7.2	-9.7	1.1	-13.3
114	Grok Build 0.1x-ai/grok-build-0.1	SpaceXAI	28.9	$1.00 / $2.00	5 of 8 benchmarks				-9.1	-8.6	-5.4	1.0	-21.4
115	Grok 4.3x-ai/grok-4.3	SpaceXAI	27.4	$1.25 / $2.50	5 of 8 benchmarks				-14.7	-5.7	-10.8	1.1	-41.8
116	Gemini 3 Flash Previewgoogle/gemini-3-flash-preview	Google	26.2	$0.50 / $3.00	6 of 8 benchmarks			62.0%	-8.5	-3.4	-7.2	-0.7	-21.0
117	Solar Pro 4upstage/solar-pro4	Upstage	24.9	$0.03 / $0.12	5 of 8 benchmarks				-10.1	-13.3	-8.4	0.5	-13.6
118	MiniMax M2.7minimax/minimax-m2.7	MiniMax	24.4	$0.30 / $1.20	5 of 8 benchmarks				-11.2	-13.7	-10.9	1.0	-16.8
119	Mistral Medium 3.5mistralai/mistral-medium-3-5	Mistral	23.8	$1.50 / $7.50	5 of 8 benchmarks				-7.0	-11.0	-9.7	-2.9	-1.7
120	Gemma 4 31Bgoogle/gemma-4-31b-it	Google	22.7	$0.10 / $0.34	5 of 8 benchmarks				-18.9	-9.4	0.3	-31.2	-51.5
121	Gemini 3.5 Flash Litegoogle/gemini-3.5-flash-lite	Google	21.5	$0.30 / $2.50	5 of 8 benchmarks				-10.4	-10.3	-12.7	-0.3	-15.1
122	Nemotron 3 Ultranvidia/nemotron-3-ultra-550b-a55b	NVIDIA	20.3	$0.60 / $3.60	5 of 8 benchmarks				-14.6	-19.9	-15.4	0.5	-24.1

How this ranks

Every benchmark value becomes a percentile among the models that have it, so accuracy scores, Elo ratings and word error rates compare without hand-tuned scaling. Metrics where lower is better are inverted first. Raw values are never summed or averaged across benchmarks. A model's mean percentile is then shrunk toward the mean of the models that were broadly benchmarked, so a model tested twice cannot outrank a broadly tested one on two lucky results. Turning a data source off runs that same ranking code again in your browser over the sources you left on.

A model scored on fewer than 3 of the 8 ranked benchmarks in this category still ranks here, on the benchmarks it does have, and its row carries a partial coverage mark. On an equal score it sits under the model that earned the same number across more of the board.

Data sources

Turn a source off to drop every benchmark it feeds and rank the board again from what is left, in your browser. Turn them all off and the table has nothing to rank. Your choice follows you across the leaderboard pages.

LMArenaCC BY 4.0
Arena ratings by LMArena, from the public leaderboard dataset.
MCP AtlasMIT
All-1,000-task Pass Rate from Scale's current Performance Comparison, using the standardized MCP loop and 100-tool-call budget.
SWE-bench
Resolve rates published by the SWE-bench maintainers.
Terminal-Bench 2.1Apache-2.0
Published Accuracy over 89 Terminal-Bench 2.1 tasks, run and verified by the benchmark team; results are model plus agent harness plus effort, not model-only evaluations.