Z.ai: GLM 5.3 Flash
z-ai/glm-5.3-flash
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while reducing compute overhead.
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), or Floor (cheapest).
| $0.0500 | $0.4500 | $0.0250 | 0.74s | 139 tok/s | ||
| $0.0399 | $0.0784 | $0.0300 | 0.83s | 79 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.74s | 141 tok/s | ||
| $0.1300 | $0.5000 | $0.0400 | 0.56s | 138 tok/s | ||
| $0.1200 | $0.3800 | $0.0200 | 0.61s | 100 tok/s | ||
| $0.1080 | $0.3500 | $0.0350 | 0.38s | 113 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.92s | 134 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.95s | 120 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.42s | 106 tok/s | ||
| $0.1400 | $0.2400 | $0.0300 | 0.46s | 93 tok/s | ||
| $0.1080 | $0.3500 | $0.0350 | 0.50s | 99 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.43s | 133 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.70s | 95 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.93s | 121 tok/s | ||
| $0.0650 | $0.2400 | $0.0640 | 0.50s | 95 tok/s | ||
| $0.1200 | $0.4000 | $0.0240 | 0.97s | 71 tok/s | ||
| $0.1500 | $0.5000 | $0.0400 | 0.65s | 87 tok/s | ||
| $0.2250 | $0.7500 | $0.0450 | 0.93s | 136 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 1.48s | 65 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.49s | 60 tok/s | ||
| $0.1125 | $0.3750 | $0.0300 | 0.91s | 86 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 0.51s | 59 tok/s | ||
44% off | $0.15$0.0840 | $0.50$0.2800 | $0.03$0.0168 | 1.85s | 38 tok/s | |
| $0.1500 | $0.5000 | $0.0300 | 1.90s | 50 tok/s | ||
50% off | $0.15$0.0750 | $0.50$0.2500 | $0.03$0.0150 | 1.88s | 38 tok/s | |
41% off | $0.15$0.0885 | $0.50$0.2950 | $0.03$0.0177 | 2.96s | 42 tok/s | |
| $0.1750 | $0.6250 | $0.0225 | 1.56s | 35 tok/s | ||
| $0.1000 | $0.4000 | $0.0300 | 1.50s | 40 tok/s | ||
| $0.1750 | $0.6250 | $0.0225 | 3.32s | 29 tok/s | ||
| $0.1200 | $0.3000 | $0.0300 | 1.00s | 18 tok/s | ||
| $0.0420 | $0.1840 | $0.0410 | 0.41s | 86 tok/s | ||
Fast | $0.1875 | $0.6250 | $0.0375 | 0.81s | 148 tok/s | |
| $0.1500 | $0.5000 | $0.0300 | 0.73s | 69 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 1.84s | 30 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 1.83s | 48 tok/s | ||
| $0.1030 | $0.2980 | $0.0520 | 4.41s | 8 tok/s | ||
| $0.0750 | $0.2500 | $0.0150 | 2.80s | 43 tok/s | ||
| $0.0800 | $0.5000 | $0.0120 | 1.78s | 17 tok/s | ||
| $0.1000 | $0.5000 | $0.0300 | 2.62s | 28 tok/s | ||
| $0.1500 | $0.5000 | $0.0300 | 4.11s | 56 tok/s | ||
| $0.0650 | $0.1800 | $0.0120 | 6.85s | 9 tok/s | ||
Not used in Standard routing:Why these endpoints are not used | ||||||
| $0.0650 | $0.2500 | $0.0200 | 0.52s | 111 tok/s | ||
- Input $0.05 / Output $0.45139 tok/s99.74% uptime
- Input $0.0399 / Output $0.078479 tok/s99.97% uptime
Input $0.15 / Output $0.50141 tok/s99.69% uptime- Input $0.13 / Output $0.50138 tok/s99.57% uptime
- Input $0.12 / Output $0.38100 tok/s99.54% uptime
- Input $0.108 / Output $0.35113 tok/s100.00% uptime
Input $0.15 / Output $0.50134 tok/s100.00% uptime- Input $0.15 / Output $0.50120 tok/s100.00% uptime
- Input $0.15 / Output $0.50106 tok/s100.00% uptime
- Input $0.14 / Output $0.2493 tok/s99.77% uptime
- Input $0.108 / Output $0.3599 tok/s100.00% uptime
- Input $0.15 / Output $0.50133 tok/s99.92% uptime
- Input $0.15 / Output $0.5095 tok/s100.00% uptime
- Input $0.15 / Output $0.50121 tok/s99.71% uptime
- Input $0.065 / Output $0.2495 tok/s100.00% uptime
- Input $0.12 / Output $0.4071 tok/s99.89% uptime
- Input $0.15 / Output $0.5087 tok/s99.97% uptime
Input $0.225 / Output $0.75136 tok/s99.72% uptime
Input $0.15 / Output $0.5065 tok/s99.94% uptime- Input $0.15 / Output $0.5060 tok/s99.94% uptime
- Input $0.1125 / Output $0.37586 tok/s100.00% uptime
- Input $0.15 / Output $0.5059 tok/s100.00% uptime
- Input $0.084 / Output $0.2844% off38 tok/s100.00% uptime
- Input $0.15 / Output $0.5050 tok/s100.00% uptime
- Input $0.075 / Output $0.2550% off38 tok/s99.64% uptime
- Input $0.0885 / Output $0.29541% off42 tok/s97.64% uptime
- Input $0.175 / Output $0.62535 tok/s100.00% uptime
- Input $0.10 / Output $0.4040 tok/s94.25% uptime
- Input $0.175 / Output $0.62529 tok/s99.91% uptime
- Input $0.12 / Output $0.3018 tok/s100.00% uptime
- Input $0.042 / Output $0.18486 tok/s100.00% uptime
- Input $0.1875 / Output $0.625148 tok/s99.89% uptimeFast
- Input $0.15 / Output $0.5069 tok/s100.00% uptime
- Input $0.15 / Output $0.5030 tok/s92.79% uptime
- Input $0.15 / Output $0.5048 tok/s99.94% uptime
- Input $0.103 / Output $0.2988 tok/s99.08% uptime
- Input $0.075 / Output $0.2543 tok/s99.53% uptime
Input $0.08 / Output $0.5017 tok/s99.92% uptime- Input $0.10 / Output $0.5028 tok/s99.28% uptime
Input $0.15 / Output $0.5056 tok/s99.89% uptime
Input $0.065 / Output $0.189 tok/s93.97% uptime- Not used in Standard routing:Why these endpoints are not used
- Input $0.065 / Output $0.25111 tok/s99.25% uptimeFlex
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.
Weighted average
GLM 5.3 Flash — Price history
| $0.0282 | $0.4986 | 92.0% | 9.2% | ||
| $0.0314 | $0.0779 | 85.8% | 5.5% | ||
| $0.0322 | $0.2502 | 71.8% | 0.7% | ||
| $0.0378 | $0.3046 | 75.9% | 2.7% | ||
| $0.0401 | $0.3797 | 79.9% | 14.5% | ||
| $0.0412 | $0.4996 | 90.7% | 1.6% | ||
| $0.0412 | $0.1785 | 44.8% | 0.0% | ||
| $0.0413 | $0.1833 | 64.7% | 3.6% | ||
| $0.0420 | $0.2941 | 65.6% | 0.5% | ||
| $0.0422 | $0.4996 | 55.5% | 1.2% | ||
| $0.0435 | $0.3494 | 88.3% | 4.5% | ||
| $0.0445 | $0.3494 | 87.0% | 2.1% | ||
| $0.0471 | $0.4997 | 85.8% | 0.3% | ||
| $0.0472 | $0.4997 | 85.7% | 2.4% | ||
| $0.0478 | $0.4998 | 85.1% | 5.1% | ||
| $0.0478 | $0.4995 | 85.1% | 0.2% | ||
| $0.0483 | $0.4998 | 84.8% | 10.6% | ||
| $0.0508 | $0.3996 | 70.3% | 1.4% | ||
| $0.0509 | $0.4999 | 70.2% | 1.7% | ||
| $0.0536 | $0.3748 | 71.4% | 4.9% | ||
| $0.0545 | $0.3996 | 68.2% | 0.3% | ||
| $0.0554 | $0.4996 | 82.8% | 2.7% | ||
| $0.0565 | $0.2993 | 70.5% | 0.4% | ||
| $0.0573 | $0.2393 | 75.2% | 0.7% | ||
| $0.0574 | $0.4997 | 77.2% | 0.8% | ||
| $0.0577 | $0.2495 | 28.8% | 0.1% | ||
| $0.0585 | $0.4997 | 76.3% | 1.1% | ||
| $0.0585 | $0.2792 | 78.6% | 0.7% | ||
| $0.0585 | $0.4996 | 76.2% | 0.9% | ||
| $0.0598 | $0.4999 | 75.2% | 8.0% | ||
| $0.0643 | $0.2394 | 65.3% | 0.6% | ||
| $0.0647 | $0.4997 | 71.0% | 2.5% | ||
| $0.0659 | $0.4724 | 75.3% | 1.0% | ||
| $0.0828 | $0.4997 | 55.9% | 0.0% | ||
| $0.0838 | $0.4999 | 55.2% | 0.0% | ||
| $0.0841 | $0.4994 | 54.9% | 0.5% | ||
| $0.0865 | $0.6244 | 58.0% | 0.0% | ||
| $0.0899 | $0.4998 | 50.1% | 2.3% | ||
Fast | $0.1009 | $0.6246 | 57.7% | 0.4% | |
| $0.1299 | $0.6246 | 29.5% | 0.0% | ||
| $0.1316 | $0.7496 | 51.9% | 3.7% | ||
Not used in Standard routing:Why these endpoints are not used | |||||
| $0.0371 | $0.2496 | 61.9% | 0.3% | ||
- Effective in $0.0282 / Effective out $0.4986Token share9.2%
- Effective in $0.0314 / Effective out $0.0779Token share5.5%
- Effective in $0.0322 / Effective out $0.2502Token share0.7%
- Effective in $0.0378 / Effective out $0.3046Token share2.7%
- Effective in $0.0401 / Effective out $0.3797Token share14.5%
- Effective in $0.0412 / Effective out $0.4996Token share1.6%
- Effective in $0.0412 / Effective out $0.1785Token share0.0%
- Effective in $0.0413 / Effective out $0.1833Token share3.6%
- Effective in $0.0420 / Effective out $0.2941Token share0.5%
- Effective in $0.0422 / Effective out $0.4996Token share1.2%
- Effective in $0.0435 / Effective out $0.3494Token share4.5%
- Effective in $0.0445 / Effective out $0.3494Token share2.1%
- Effective in $0.0471 / Effective out $0.4997Token share0.3%
- Effective in $0.0472 / Effective out $0.4997Token share2.4%
- Effective in $0.0478 / Effective out $0.4998Token share5.1%
- Effective in $0.0478 / Effective out $0.4995Token share0.2%
- Effective in $0.0483 / Effective out $0.4998Token share10.6%
- Effective in $0.0508 / Effective out $0.3996Token share1.4%
- Effective in $0.0509 / Effective out $0.4999Token share1.7%
- Effective in $0.0536 / Effective out $0.3748Token share4.9%
- Effective in $0.0545 / Effective out $0.3996Token share0.3%
- Effective in $0.0554 / Effective out $0.4996Token share2.7%
- Effective in $0.0565 / Effective out $0.2993Token share0.4%
- Effective in $0.0573 / Effective out $0.2393Token share0.7%
- Effective in $0.0574 / Effective out $0.4997Token share0.8%
- Effective in $0.0577 / Effective out $0.2495Token share0.1%
- Effective in $0.0585 / Effective out $0.4997Token share1.1%
- Effective in $0.0585 / Effective out $0.2792Token share0.7%
- Effective in $0.0585 / Effective out $0.4996Token share0.9%
- Effective in $0.0598 / Effective out $0.4999Token share8.0%
- Effective in $0.0643 / Effective out $0.2394Token share0.6%
- Effective in $0.0647 / Effective out $0.4997Token share2.5%
- Effective in $0.0659 / Effective out $0.4724Token share1.0%
- Effective in $0.0828 / Effective out $0.4997Token share0.0%
- Effective in $0.0838 / Effective out $0.4999Token share0.0%
- Effective in $0.0841 / Effective out $0.4994Token share0.5%
- Effective in $0.0865 / Effective out $0.6244Token share0.0%
- Effective in $0.0899 / Effective out $0.4998Token share2.3%
- Effective in $0.1009 / Effective out $0.6246FastToken share0.4%
- Effective in $0.1299 / Effective out $0.6246Token share0.0%
- Effective in $0.1316 / Effective out $0.7496Token share3.7%
- Not used in Standard routing:Why these endpoints are not used
- Effective in $0.0371 / Effective out $0.2496FlexToken share0.3%
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).
P50, best across providers
P50, best provider
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.
Availability over the last 3 days
Last 72 hoursAvailability over the last 24 hours
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.
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.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | GLM-5.3-Flash Intelligence Index | 27.8 |
| Artificial Analysis | GLM-5.3-Flash Finance & Accounting Index | 38.8 |
| Artificial Analysis | GLM-5.3-Flash Strategy & Ops Index | 44.5 |
| Artificial Analysis | GLM-5.3-Flash Legal Index | 34.8 |
| Artificial Analysis | GLM-5.3-Flash Healthcare & Medical Index | 36.6 |
| Artificial Analysis | GLM-5.3-Flash Engineering Index | 34.5 |
| Artificial Analysis | GLM-5.3-Flash Economics Index | 35.3 |
| Artificial Analysis | GLM-5.3-Flash Cyber Index | 49.9 |
| Artificial Analysis | GLM-5.3-Flash Openness Index | 44.4 |
| Artificial Analysis | GLM-5.3-Flash AutomationBench-AA | 60.4% |
| Artificial Analysis | GLM-5.3-Flash GDP.pdf | 15.4% |
| Artificial Analysis | GLM-5.3-Flash AA-AnalystAgent | 25.0% |
| Artificial Analysis | GLM-5.3-Flash EnterpriseOps-Gym-AA | 33.2% |
| Artificial Analysis | GLM-5.3-Flash ITBench-AA | 51.2% |
| Artificial Analysis | GLM-5.3-Flash Harvey LAB-AA | 0.8% |
| Artificial Analysis | GLM-5.3-Flash GPQA Diamond | 91.2% |
| Artificial Analysis | GLM-5.3-Flash HLE | 39.9% |
| Artificial Analysis | GLM-5.3-Flash AA-LCR | 80.0% |
| Artificial Analysis | GLM-5.3-Flash τ³-Banking | 47.2% |
| Artificial Analysis | GLM-5.3-Flash GDPval-AA | 57.4% |
| Artificial Analysis | GLM-5.3-Flash CritPt | 15.4% |
| Artificial Analysis | GLM-5.3-Flash HLE-Diamond | 18.2% |
| Artificial Analysis | GLM-5.3-Flash SciCode | 51.6% |
| Artificial Analysis | GLM-5.3-Flash Terminal-Bench 4.0 | 32.8% |
| Artificial Analysis | GLM-5.3-Flash Terminal-Bench 2.1 | 84.3% |
| Artificial Analysis | GLM-5.3-Flash FrontierCode-AA 2.0 | 17.6% |
| Artificial Analysis | GLM-5.3-Flash FrontierCode 1.1 | 41.0% |
| Artificial Analysis | GLM-5.3-Flash Terminal-Bench-Science | 4.8% |
| Artificial Analysis | GLM-5.3-Flash CritPt v2 (Hard Subset) | 24.5% |
| Artificial Analysis | GLM-5.3-Flash CritPt v2 | 46.3% |
| Artificial Analysis | GLM-5.3-Flash AA-Omniscience Accuracy | 28.2% |
| Artificial Analysis | GLM-5.3-Flash AA-Omniscience Non-Hallucination Rate | 61.7% |
| Artificial Analysis | GLM-5.3-Flash CWE-Bench-AA | 55.8% |
| Artificial Analysis | GLM-5.3-Flash DeepsecBench-AA | 19.9% |
| Artificial Analysis | GLM-5.3-Flash CyberGym-E2E-AA | 74.0% |
| Artificial Analysis | GLM-5.3-Flash MLCR-AA | 11.7% |
| Artificial Analysis | GLM-5.3-Flash VIALS-AA | 39.8% |
| Design Arena | GLM-5.3-Flash Agents Arena Pdf Elo | 1136 |
| Design Arena | GLM-5.3-Flash Models Arena 3D Elo | 1286 |
| Design Arena | GLM-5.3-Flash Models Arena Asciiart Elo | 1255 |
| Design Arena | GLM-5.3-Flash Models Arena Code Categories Elo | 1259 |
| Design Arena | GLM-5.3-Flash Models Arena Data Visualization Elo | 1256 |
| Design Arena | GLM-5.3-Flash Models Arena Game Development Elo | 1260 |
| Design Arena | GLM-5.3-Flash Models Arena SVG Elo | 1266 |
| Design Arena | GLM-5.3-Flash Models Arena UI Component Elo | 1293 |
| Design Arena | GLM-5.3-Flash Models Arena Website Elo | 1253 |
| Vals AI | GLM 5.3 Flash Access to Justice Benchmark | 71.1% |
| Vals AI | GLM 5.3 Flash Code Migration v1.1 | 51.6% |
| Vals AI | GLM 5.3 Flash CyberBench v1.1 | 67.9% |
| Vals AI | GLM 5.3 Flash CyberOffense | 18.3% |
| Vals AI | GLM 5.3 Flash EMB v1.1 | 66.3% |
| Vals AI | GLM 5.3 Flash Finance Agent v2.1 | 60.6% |
| Vals AI | GLM 5.3 Flash Vals Finance Index | 55.8% |
| Vals AI | GLM 5.3 Flash GPQA Diamond | 86.4% |
| Vals AI | GLM 5.3 Flash HLAB-Vals-1.1 | 10.4% |
| Vals AI | GLM 5.3 Flash IOI | 52.5% |
| Vals AI | GLM 5.3 Flash LiveCodeBench | 80.5% |
| Vals AI | GLM 5.3 Flash LegalBench | 83.9% |
| Vals AI | GLM 5.3 Flash Vals Legal Index | 71.1% |
| Vals AI | GLM 5.3 Flash Legal Research Bench | 45.2% |
| Vals AI | GLM 5.3 Flash MedScribe | 88.9% |
| Vals AI | GLM 5.3 Flash MMLU Pro | 86.1% |
| Vals AI | GLM 5.3 Flash MMMU Pro | 86.0% |
| Vals AI | GLM 5.3 Flash ProgramBench | 0.0% |
| Vals AI | GLM 5.3 Flash ProofBench v1.1 | 21.0% |
| Vals AI | GLM 5.3 Flash SAGE | 89.3% |
| Vals AI | GLM 5.3 Flash SkillsBench | 40.2% |
| Vals AI | GLM 5.3 Flash SRE Bench | 0.8% |
| Vals AI | GLM 5.3 Flash SWE-bench | 92.0% |
| Vals AI | GLM 5.3 Flash Tax Agent Bench | 31.6% |
| Vals AI | GLM 5.3 Flash TaxEval v2 | 75.6% |
| Vals AI | GLM 5.3 Flash Terminal-Bench 4.0 | 37.4% |
| Vals AI | GLM 5.3 Flash Terminal-Bench-Science | 2.9% |
| Vals AI | GLM 5.3 Flash Vals Index | 50.9% |
| Vals AI | GLM 5.3 Flash Vibe Code Bench 1-100 | 16.0% |
| Vals AI | GLM 5.3 Flash Vibe Code Bench v1.1 | 30.8% |
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.
Available through EU in-region routing
Z.ai: GLM 5.3 Flash is available through EU in-region routing. Requests sent to eu.orr.api.report.finance are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).
- Providers in Europe
- Inceptron
- Prices in Europe
- Inceptron: $0.15/M tokens input, $0.50/M tokens output
- Base URL
https://eu.orr.api.report.finance/api/v1- Plan
- Requires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing
Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs
Documentation
Guides, cookbooks, blog posts, and tools for building with Z.ai: GLM 5.3 Flash on OpenRouter.
Frequently asked questions
Yes, the stealth model Ox Alpha was revealed to be ZAI's new model, GLM-5.3 Flash.
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while reducing compute overhead.
The GLM 5.3 Flash documentation on OpenRouter includes the Open-Weight vs Closed-Weight Models: What's the Difference? blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.
GLM 5.3 Flash costs $0.0399/M input tokens and $0.0784/M output tokens, with separate rates for Cache Read at $0.03/M tokens.
GLM 5.3 Flash has a 1,048,576 token context window. It supports up to 131,072 completion tokens.
Yes. GLM 5.3 Flash accepts tools for function calling. It also supports structured outputs via a JSON schema in response_format.
GLM 5.3 Flash accepts text, images, and video as input and returns text.
GLM 5.3 Flash is served by 36 providers on OpenRouter: Relace, Wafer, Open Inference, Darkbloom, Particle AI, Tenstorrent, GMICloud, NEAR AI and 28 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 Flash was released on August 26, 2026.