Skip to content
Alpha
  • Models
  • Rankings
  • Ori
ElevenLabs launch offer: every ElevenLabs model is 50% off through October 19, 2026. See ElevenLabs models

Footer

OpenRouterOpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Benchmarks
  • Apps
  • Discover
  • Models
  • Ori
  • Collections
  • Providers
  • Tools
  • Pricing
  • Business
  • Enterprise
  • Labs

Company

  • About
  • Blog
  • Careers
    Hiring
  • Privacy
  • Terms of Service
  • Trust Center
  • Support
  • Works With OR
  • Data
  • Brand

Developer

  • Documentation
  • API Reference
  • Developer Platform
  • Status
  • AI Site Map

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube
Favicon for z-ai

Z.ai: GLM 5.3

z-ai/glm-5.3

Model weights
Compare

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves on GLM-5.2 in coding and in the balance between performance and token efficiency.

Reasoning is always on and cannot be disabled. Reasoning efforts low, high, and max are supported; max is the default.

Modalities
In / Out Price
$0.0608 / $4.40per 1M
Context
1.0M
Released
Aug 18, 2026
Compare
ProvidersPricingPerformanceUptimeBenchmarksAppsActivityFAQExplore

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), Floor (cheapest), or Exacto (highest tool-calling accuracy).

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Z.ai: GLM 5.3 (Artificial Analysis, Design Arena and Vals AI)
SourceBenchmarkScore
Artificial AnalysisGLM-5.3 (Max) Intelligence Index44.8
Artificial AnalysisGLM-5.3 (Max) Coding Index74.8
Artificial AnalysisGLM-5.3 (Max) Agentic Index53.1
Artificial AnalysisGLM-5.3 (Max) GPQA Diamond91.7%
Artificial AnalysisGLM-5.3 (Max) HLE42.3%
Artificial AnalysisGLM-5.3 (Max) AA-LCR79.7%
Artificial AnalysisGLM-5.3 (Max) τ-Bench Banking50.3%
Artificial AnalysisGLM-5.3 (Max) GDPval-AA57.5%
Artificial AnalysisGLM-5.3 (Max) CritPt19.1%
Artificial AnalysisGLM-5.3 (Max) SciCode59.0%
Artificial AnalysisGLM-5.3 (Max) Terminal-Bench 4.041.9%
Artificial AnalysisGLM-5.3 (Max) Terminal-Bench 2.183.9%
Artificial AnalysisGLM-5.3 (Max) AA-Omniscience Accuracy33.9%
Artificial AnalysisGLM-5.3 (Max) AA-Omniscience Non-Hallucination Rate70.4%
Artificial AnalysisGLM-5.3 (Low) Intelligence Index34.3
Artificial AnalysisGLM-5.3 (Low) HLE36.6%
Artificial AnalysisGLM-5.3 (Low) AA-LCR72.7%
Artificial AnalysisGLM-5.3 (Low) GDPval-AA41.1%
Artificial AnalysisGLM-5.3 (Low) CritPt14.6%
Artificial AnalysisGLM-5.3 (Low) SciCode42.0%
Artificial AnalysisGLM-5.3 (Low) Terminal-Bench 4.034.8%
Artificial AnalysisGLM-5.3 (Low) AA-Omniscience Accuracy33.9%
Artificial AnalysisGLM-5.3 (Low) AA-Omniscience Non-Hallucination Rate36.0%
Design ArenaGLM-5.3 Models Arena 3D Elo1333
Design ArenaGLM-5.3 Models Arena Asciiart Elo1213
Design ArenaGLM-5.3 Models Arena Code Categories Elo1307
Design ArenaGLM-5.3 Models Arena Data Visualization Elo1260
Design ArenaGLM-5.3 Models Arena Game Development Elo1315
Design ArenaGLM-5.3 Models Arena SVG Elo1291
Design ArenaGLM-5.3 Models Arena UI Component Elo1314
Design ArenaGLM-5.3 Models Arena Website Elo1298
Design ArenaGLM-5.3 Agents Arena Htmlslides Elo1189
Design ArenaGLM-5.3 Agents Arena Mobile Apps Elo1191
Design ArenaGLM-5.3 Agents Arena Python-Pptxslides Elo1255
Vals AIGLM 5.3 Code Migration44.2%
Vals AIGLM 5.3 CyberBench v1.172.1%
Vals AIGLM 5.3 EMB56.3%
Vals AIGLM 5.3 Finance Agent (v2)55.8%
Vals AIGLM 5.3 GPQA Diamond88.1%
Vals AIGLM 5.3 Harvey's Legal Agent Benchmark8.3%
Vals AIGLM 5.3 IOI68.4%
Vals AIGLM 5.3 LiveCodeBench80.5%
Vals AIGLM 5.3 LegalBench84.8%
Vals AIGLM 5.3 Legal Research Bench49.0%
Vals AIGLM 5.3 MedCode42.9%
Vals AIGLM 5.3 MedScribe88.8%
Vals AIGLM 5.3 MMLU Pro86.8%
Vals AIGLM 5.3 ProgramBench1.5%
Vals AIGLM 5.3 ProofBench v1.149.0%
Vals AIGLM 5.3 Vals RSI Index22.9%
Vals AIGLM 5.3 SkillsBench47.5%
Vals AIGLM 5.3 SRE Bench9.2%
Vals AIGLM 5.3 SWE-bench95.4%
Vals AIGLM 5.3 Tax Agent Bench39.7%
Vals AIGLM 5.3 TaxEval v272.4%
Vals AIGLM 5.3 Terminal-Bench 4.038.9%
Vals AIGLM 5.3 Vals Index53.5%
Vals AIGLM 5.3 Vibe Code Bench 1-10020.0%
Vals AIGLM 5.3 Vibe Code Bench v1.178.1%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Z.ai: GLM versions

Explore more models

AI Model RankingsRanking
$0.0608$4.400$0.05651.40s61 tps
100.00%
$0.0800$4.800$0.06000.72s102 tps
99.95%
$0.1100$6.000$0.10900.75s83 tps
99.78%
$0.1190$6.000$0.11901.68s34 tps
99.75%
30% off
$0.20$0.1400$6.00$4.200$0.15$0.10502.17s17 tps
99.49%
$0.1400$4.400$0.10000.96s126 tps
94.04%
$0.1500$3.400$0.14002.42s53 tps
99.68%
$0.1500$3.400$0.14002.10s61 tps
99.67%
$0.1900$4.400$0.19001.48s87 tps
99.96%
$0.3000$6.000$0.29900.92s71 tps
99.68%
72% off
$1.40$0.3990$4.40$1.254$0.26$0.07411.08s62 tps
99.93%
$0.4900$2.900$0.19002.20s44 tps
99.28%
$0.5500$1.254$0.18000.72s103 tps
99.80%
38% off
$0.90$0.5625$4.00$2.500$0.20$0.12501.15s69 tps
98.31%
50% off
$1.40$0.7000$4.40$2.200$0.26$0.13001.86s56 tps
99.73%
50% off
$1.40$0.7000$4.40$2.200$0.26$0.13002.51s57 tps
99.72%
40% off
$1.40$0.8400$4.40$2.640$0.26$0.15602.79s58 tps
99.54%
$0.8400$2.660$0.19001.16s116 tps
99.95%
$0.9100$2.860$0.16900.71s93 tps
100.00%
30% off
$1.40$0.9800$4.40$3.080$0.26$0.18202.81s36 tps
100.00%
$1.1900$3.740$0.23801.71s61 tps
99.86%
10% off
$1.40$1.2600$4.40$3.960$0.26$0.23400.91s93 tps
100.00%
$1.4000$4.400$0.14001.26s104 tps
99.57%
$1.4000$4.400$0.14003.02s50 tps
99.94%
$1.4000$4.400$0.14001.58s100 tps
99.48%
$1.4000$4.400$0.13001.22s97 tps
99.78%
$1.4000$4.400$0.26001.07s106 tps
100.00%
20% off
$1.75$1.4000$5.50$4.400$0.325$0.26001.37s47 tps
96.34%
$1.4000$4.400$0.26001.26s62 tps
98.73%
$1.4000$4.400$0.26000.93s68 tps
99.67%
$1.4000$4.400$0.26001.18s65 tps
100.00%
$1.4000$4.400$0.14003.43s51 tps
100.00%
$1.4000$4.400$0.26000.81s58 tps
98.59%
$1.4000$4.400$0.26003.21s28 tps
97.07%
$1.4000$4.400$0.26001.73s53 tps
99.24%
$1.4000$4.400$0.26002.86s57 tps
100.00%
$2.1000$6.600$0.40000.32s156 tps
78.30%
Fast
$2.1000$6.600$0.39000.85s147 tps
98.85%
Fast
$2.1000$6.600$0.21000.71s151 tps
99.96%
Fast
$2.1000$6.600$0.21000.65s167 tps
100.00%
Prime
$2.8000$8.800$0.56002.25s46 tps
100.00%
Not used in Standard routing:Why these endpoints are not used
$1.5400$4.840$0.15400.43s100 tps
100.00%
Throughput
167tok/s

P50, best across providers

Latency
0.32s

P50, best provider

AutoExacto benchmarks
GPQA DiamondTAU-BenchOpenInference87.9%--Baidu Qianfan86.5%--Baseten (fp4)86.3%--DeepInfra88.3%79.1%Z.ai88.8%77.1%Baseten (fp4) (US)86.3%78.1%Baseten (fp8) (US)83.3%80.0%Baseten (fp8)84.2%78.9%Baidu Qianfan85.5%75.8%
+32 more providers
Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.88%

Availability over the last 3 days

Last 72 hours
Availability 99.88%
3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability
99.88%
Without Routing
95.64%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

1.
Favicon for https://claude.ai/apple-touch-icon.png
Claude Code
Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.
2.55Ttokens
2.
Favicon for https://strix.ai/
Strix
new
1Ttokens
3.
Favicon for https://nousresearch.com
Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
800Btokens
4.
Favicon for https://omp.sh/
omp
new
500Btokens
5.
Favicon for https://pi.dev/
pi
There are many coding agents, but this one is yours.
495Btokens

Frequently asked questions

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves on GLM-5.2 in coding and in the balance between performance and token efficiency. Reasoning is always on and cannot be disabled. Reasoning efforts low, high, and max are supported; max is the default.

GLM 5.3 costs $0.0608/M input tokens and $4.40/M output tokens, with separate rates for Cache Read at $0.0565/M tokens.

GLM 5.3 has a 1,048,576 token context window. It supports up to 131,072 completion tokens.

Yes. GLM 5.3 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

GLM 5.3 is served by 33 providers on OpenRouter: Relace, inference.net, Wafer, Morph, Reka AI, Makora, Sail Research, AkashML and 25 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

GLM 5.3 was released on August 18, 2026.