AA-Briefcase v1.1

Une évaluation privée des capacités agentiques de pointe en travail de connaissance sur longue durée, testant les agents sur des flux de travail professionnels réalistes qui exigent des livrables tels que des tableurs, des présentations et des notes de synthèse.

Poids de l'Indice d'intelligence: 15%
Modèles: 152 actuel, 263 au total
Médiane: 1186
Plage: 0–1816

Distribution des scores

Le plus bas: 0
Médiane: 1186
Le plus élevé: 1816

Classement

Inclure les modèles obsolètes

Rang	Modèle	Score	Coût / tâche	Temps / tâche	Tokens de sortie / tâche
1	Claude Sonnet 5.5 (Max, Default Fallback) Anthropic	1816	$20.47	67m 51s	625K
2	Claude Opus 5.5 (Max, Default Fallback) Anthropic	1801	$21.05	48m 13s	338K
3	Claude Fable 5.2 (Max, Default Fallback) Anthropic	1785	$59.35	89m 17s	550K
4	Claude Opus 5.5 (Xhigh, Default Fallback) Anthropic	1760	$12.27	30m 27s	198K
5	Claude Sonnet 5.5 (Xhigh, Default Fallback) Anthropic	1742	$6.98	28m 59s	242K
6	Claude Fable 5.2 (Xhigh, Default Fallback) Anthropic	1738	$28.56	46m 17s	287K
7	Muse Wonder Full (Max) Meta	1729	—	—	107K
8	Claude Opus 5.5 (High, Default Fallback) Anthropic	1682	$6.27	17m 59s	113K
9	Claude Fable 5.2 (High, Default Fallback) Anthropic	1677	$15.49	30m 6s	168K
10	Kimi K3.1 Pro Kimi	1674	$7.22	74m 7s	144K
11	Muse Wonder (Max) Meta	1659	$8.68	15m 17s	112K
12	Grok 4.7 (Xhigh) SpaceXAI	1636	$8.03	42m 1s	161K
13	Claude Sonnet 5.5 (High, Default Fallback) Anthropic	1634	$3.27	16m 33s	140K
14	Grok 4.7 (High) SpaceXAI	1629	$6.38	34m 2s	137K
15	Claude Fable 5.2 (Medium, Default Fallback) Anthropic	1622	$6.84	14m 45s	80.3K
16	Claude Opus 5.5 (Medium, Default Fallback) Anthropic	1620	$4.40	13m 28s	83.4K
17	Gemini 4 Carbon (High) Google	1611	$8.16	27m 58s	138K
18	Muse Wonder (High) Meta	1610	$5.67	11m 12s	81.5K
19	Qwen3.8 Max (0902) Alibaba	1605	$19.58	109m 4s	334K
20	Gemini 4 Carbon (Medium) Google	1602	$7.06	27m 34s	122K
21	Claude Haiku 5.5 (Max) Anthropic	1573	$0.67	27m 53s	425K
22	Qwen3.8-Flash-Next Alibaba	1573	—	—	—
23	DeepSeek V4.1 Flash 1017 (Max) DeepSeek	1570	$0.68	24m 20s	202K
24	GPT-6 Astra (Max) OpenAI	1569	$9.50	30m 26s	68.7K
25	Gemini 4 Carbon (Low) Google	1567	$7.20	22m 27s	86.9K
26	Claude Fable 5.2 (Low, Default Fallback) Anthropic	1563	$5.62	11m 59s	66.7K
27	GPT-6.1 Sol (Max) OpenAI	1555	$2.31	34m 6s	115K
28	GPT-6 Astra (Xhigh) OpenAI	1544	$6.56	19m 9s	44.7K
29	Muse Ember (Xhigh) Meta	1527	$0.10	2m 41s	73.6K
30	Inkling 2 Small Thinking Machines	1527	$1.02	21m 13s	162K
31	Claude Haiku 5.5 (Xhigh) Anthropic	1523	$0.37	16m 55s	240K
32	Ling 3.1 Flash 1019 InclusionAI	1522	$3.88	—	350K
33	Muse Ember (Max) Meta	1521	$0.13	3m 19s	90.5K
34	MiMo-V2.6-Pro Xiaomi	1512	$0.38	48m 25s	164K
35	GPT-6 Astra (High) OpenAI	1505	$4.76	13m 37s	32.1K
36	GPT-6.1 Sol (Xhigh) OpenAI	1503	$1.04	13m 0s	43.5K
37	GLM-5.3 (Max) Z AI	1503	$5.65	33m 5s	136K
38	Agnes 2.5 Pro Theta Sapiens AI	1496	$1.34	24m 15s	281K
39	Grok 4.7 (Low) SpaceXAI	1493	$2.10	17m 43s	68.7K
40	MiMo-V2.6-Flash Xiaomi	1487	$0.15	76m 37s	161K
41	Gemini 4 Argon (High) Google	1484	$11.66	35m 0s	142K
42	MiniMax-M3.1-Flash MiniMax	1474	$0.53	16m 10s	155K
43	GPT-6.1 Sol (High) OpenAI	1469	$0.84	9m 59s	34.1K
44	GPT-6 Astra (Medium) OpenAI	1461	$3.94	10m 19s	26.4K
45	GLM-5.3-Flash Z AI	1450	$0.98	44m 25s	132K
46	Tencent Hy4 Tencent	1447	$1.31	—	89K
47	Claude Haiku 5.5 (High) Anthropic	1441	$0.22	11m 27s	157K
48	Claude Sonnet 5.5 (Medium, Default Fallback) Anthropic	1439	$1.39	8m 34s	71.6K
49	Qwen3.8 2.4T A95B Alibaba	1430	—	—	—
50	DeepSeek V4.1 Flash 1017 (Non-reasoning) DeepSeek	1425	$0.82	16m 4s	121K

1
Claude Sonnet 5.5 (Max, Default Fallback)
Anthropic
1816
$20.47
67m 51s
625K
2

Données de Artificial Analysis

Command Palette

AA-Briefcase v1.1

Distribution des scores

Classement