Passer au contenu principal

GLM-5.1 (Reasoning)

Z AIGLMOpen WeightMIT · Usage Commercial

Description

GLM-5.1 is Z.AI's next-generation flagship foundation model designed for long-horizon agentic engineering tasks. Built on a 754B MoE architecture (40B active parameters), it can work continuously and autonomously on a single task for up to 8 hours, completing the full loop from planning and execution to iterative optimization and delivery. GLM-5.1 achieves state-of-the-art on SWE-Bench Pro (58.4) and demonstrates strong performance across coding, reasoning, and agentic benchmarks. It supports 200K context length, 128K max output tokens, thinking mode, function calling, structured output, context caching, and MCP integration. Overall performance is aligned with Claude Opus 4.6 with particular strengths in sustained execution and complex engineering optimization.

Date de sortie
2026-04-07
Paramètres
754.0B
Longueur du contexte
200K
Modalités
text

Radar de capacités

38
general
54
coding
87
reasoning
60
scienceest.
60
agents
0
multimodal

Science utilise un proxy de raisonnement lorsque les benchmarks scientifiques dédiés ne sont pas disponibles.

Classements

Domaine#RangScoreSource
Capacité agentique33
60.0
LS
Classement codage60
74.0
AA
Classement général21
81.0
AA
Science43
72.0
AA

Scores de benchmarks (LLM Stats)

Agents

Vending-Bench 2563441.0%Aut.
GDPval-AA1281.00 / 3000Aut.
BrowseComp79.3%Aut.
MCP Atlas71.8%Aut.
TAU3-Bench70.6%Aut.
Terminal-Bench 2.069.0%Aut.
CyberGym68.7%Aut.
SWE-Bench Pro58.4%Aut.
Finance Agent v244.8%Aut.
NL2Repo42.7%Aut.
Toolathlon40.7%Aut.
FrontierSWE31.0%Aut.

Biology

GPQA86.2%Aut.

General

LiveBench70.2%Aut.

Math

AIME 202695.3%Aut.
HMMT 202594.0%Aut.
IMO-AnswerBench83.8%Aut.
HMMT Feb 2682.6%Aut.
Humanity's Last Exam52.3%Aut.

Indices d'évaluation AA

Coding Index
55.8
Intelligence Index
40.2
Tau2
1.0
Gpqa
0.9
Ifbench
0.8
Lcr
0.6
Terminalbench V2 1
0.6
Scicode
0.4
Terminalbench Hard
0.4
Hle
0.3
Tau Banking
0.1

Scores par catégorie LLM Stats

Legal
100
Finance
100
Agents
100
Reasoning
100
General
100
Physics
90
Biology
90
Chemistry
90
Math
80
Search
80
Safety
70
Code
60
Tool Calling
60
Vision
50
Coding
40

Tarification

Prix d'entrée$1.4 / 1M tokens
Prix de sortie$4.4 / 1M tokens
Prix mixte (3:1)$2.15 / 1M tokens
Prix de lecture cache$0.26 / 1M tokens
Prix d'écriture cacheGratuit

Vitesse

Tokens/sec99.8
Délai du premier token0.80s
Temps de réponse38.80s

Classement des Prix par Fournisseur

Classement des Prix par Fournisseur

25 fournisseurs

Moins cher: ZAIPlus cher: Merge Gateway
FournisseurEntréeSortie
1ZAIMoins cher
$0
$0
2FriendliAI
$0
$0
3NanoGPT
$0.3
$2.55
4HPC-AI
$0.615
$2.46
5ZenMux
$0.8781
$3.5126
6Lilac
$0.9
$3
7OpenRouter
$0.98
$3.08
8Hugging Face
$1
$3.2
9Wafer
$1
$3.2
10Synthetic
$1
$3
11routing.run
$1
$3
12Deep Infra
$1.05
$3.5
13FastRouter
$1.05
$3.5
14Kilo Gateway
$1.26
$3.96
15Baseten
$1.3
$4.3
16Z AIPRINCIPAL
$1.4
$4.4
17SiliconFlow (China)
$1.4
$4.4
18NovitaAI
$1.4
$4.4
19Weights & Biases
$1.4
$4.4
20Friendli
$1.4
$4.4
21SiliconFlow
$1.4
$4.4
22Vercel AI Gateway
$1.4
$4.4
23Together AI
$1.4
$4.4
24OrcaRouter
$1.4
$4.4
25Merge Gateway
$1.4
$4.4

Comparer les prix entre différents fournisseurs API pour ce modèle.

Sources externes