Passer au contenu principal
AITier

Claude Opus 4.6 (Non-reasoning, High Effort)

AnthropicClaudeProprietary

Description

Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).

Date de sortie
2026-02-05
Paramètres
—
Longueur du contexte
1.0M
Modalités
image, pdf, text

Radar de capacités

25
general
70
coding
84
reasoning
64
science
80
agents
80
multimodal

Classements

Domaine#RangScoreSource
Capacité agentique25
57.0
LS
Classement codage142
75.0
AA
Classement général187
55.0
AA
Classement multimodal25
61.0
LS
Science202
58.0
AA

Scores de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Agents

Finance Agent60.7%Aut.

Code

CyberGym73.8%Aut.
FrontierSWE56.0%
OpenRCA34.9%Aut.

Communication

Tau2 Telecom99.3%Aut.
Tau2 Retail91.9%Aut.

Language

MMMLU91.1%Aut.

Legal

Legal Agent Benchmark4.2%

Long Context

MRCR v2 (8-needle)76.0%Aut.

Math

AIME 202599.8%Aut.
LiveBench76.3%

Multimodal

OSWorld72.7%Aut.

Reasoning

Vending-Bench 2801759.0%Aut.
Graphwalks parents >128k95.4%Aut.
GPQANYU + Cohere + Anthropic (2023)91.3%Aut.
BrowseCompOpenAI (2025)84.0%Aut.
SWE-Bench Verified80.8%Aut.
SWE-bench Multilingual77.8%Aut.
CharXiv-R77.4%Aut.
ARC-AGI v268.8%Aut.
Terminal-Bench 2.0Stanford × Laude Institute (2026)65.4%Aut.
MCP Atlas62.7%Aut.
Graphwalks BFS >128k61.5%Aut.
Humanity's Last Exam53.1%Aut.

Safety

FigQA78.3%Aut.

Search

DeepSearchQA91.3%Aut.

Vision

MMMU-Pro77.3%Aut.

Indices d'évaluation AA

(Artificial Analysis)
Tau2(Sierra + U Toronto + Vector Institute (2025))
84.8
Gpqa(NYU + Cohere + Anthropic (2023))
84.0
Lcr(Artificial Analysis)
67.0
Terminalbench Hard(Stanford × Laude Institute (2026))
48.5
Ifbench(Google Research (2023))
44.6
Intelligence Index(Artificial Analysis)
26.4
Hle(Center for AI Safety + Scale AI (2025))
19.1

Scores par catégorie LLM Stats

(LLM Stats (zeroeval))
Agents
100
Reasoning
100
General
100
Communication
100
Language
90
Physics
90
Search
90
Biology
90
Chemistry
90
Long Context
80
Math
80
Multimodal
80
Safety
80
Spatial Reasoning
80
Frontend Development
80
Healthcare
80
Tool Calling
80
Code
70
Vision
70
Finance
60
Legal
0

Tarification

Prix d'entrée$5 / 1M tokens
Prix de sortie$25 / 1M tokens
Prix mixte (3:1)$10 / 1M tokens
Prix de lecture cache$0.5 / 1M tokens
Prix d'écriture cache$6.25 / 1M tokens

Vitesse

Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s

Classement des Prix par Fournisseur

Classement des Prix par Fournisseur

16 fournisseurs

Moins cher: AnthropicPlus cher: Venice AI
FournisseurEntréeSortie
1AnthropicMoins cher
$0.00001
$0.00003
2Abacus
$5
$25
3Jiekou.AI
$5
$25
4OpenCode Zen
$5
$25
5FrogBot
$5
$25
6AIHubMix
$5
$25
7Azure Cognitive Services
$5
$25
8Requesty
$5
$25
9DevPass (LLM Gateway)
$5
$25
10Azure
$5
$25
11Auriko
$5
$25
12FreeModel
$5
$25
13Neon
$5
$25
14Pioneer
$5
$25
15Opper
$5
$25
16Venice AI
$6
$30

Comparer les prix entre différents fournisseurs API pour ce modèle.

Sources externes

LLM StatsArtificial Analysis