Z.ai
GLM 5.3 Flash
Launched 2026-08-26 · Listed on OpenRouter 2026-08-26
Prices updated . Benchmarks updated .
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...
You send / 1M
$0.040
tokens in
It replies / 1M
$0.078
tokens out
Context
1M
tokens it can read at once
Typical request
$0.0001
1,000 words in, 400 out
Benchmarks
| Benchmark | Score | Kind | Date | Source |
|---|---|---|---|---|
| Epoch Capabilities Index | 153.2 | Independent | 2026-08-20 | Epoch AI |
| GPQA DiamondMax effort | 90.2% | Independent | 2026-08-26 | Epoch AI |
| OTIS Mock AIMEMax effort | 93.9% | Independent | 2026-08-26 | Epoch AI |
| ARC-AGI-2Max effort | 65.8% | Independent | 2026-08-20 | ARC Prize, republished by Epoch AI |
| GDP.pdfMax effort | 14.0% | Independent | 2026-08-20 | Surge AI GDP.pdf, republished by Epoch AI |
| CritPt | 15.4% | Independent | 2026-08-20 | CritPt, republished by Epoch AI |
| Arena Elo (text) | 1,473 | Independent | 2026-10-28 | LMArena |
| Arena Elo (webdev) | 1,607 | Independent | 2026-10-28 | LMArena |
| Arena Elo (vision) | 1,294 | Independent | 2026-10-19 | LMArena |
Rankings on this site use the Epoch Capabilities Index, an independent composite. Lab-reported rows are the lab's own figures and are kept separate.
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| Relace · fp4 | $0.040 | $0.078 | 100% | Read |
| Wafer · nvfp4 | $0.042 | $0.184 | 100% | Read |
| Morph · fp8 | $0.072 | $0.208 | 98% | Read |
| Makora · fp4 | $0.140 | $0.240 | 100% | n/a |
| Wafer · nvfp4 | $0.065 | $0.240 | 100% | Read |
| Particle AI · nvfp4 | $0.065 | $0.250 | 100% | Read |
| Tenstorrent · fp8 | $0.075 | $0.250 | 100% | Read |
| Novita · fp8 | $0.084 | $0.280 | 100% | Read |
| Reka | $0.120 | $0.300 | 100% | Read |
| River | $0.096 | $0.320 | 100% | Read |
| InferenceNet · fp4 | $0.108 | $0.350 | 100% | Read |
| InferenceNet · fp4 | $0.108 | $0.350 | 100% | Read |
| DigitalOcean | $0.113 | $0.375 | 100% | Read |
| CoreWeave · nvfp4 | $0.120 | $0.380 | 100% | Read |
| SiliconFlow · fp8 | $0.100 | $0.400 | 99% | Read |
| NextBit · fp4 | $0.120 | $0.450 | 100% | Read |
| Inceptron · fp8 | $0.175 | $0.490 | 100% | Read |
| DekaLLM | $0.100 | $0.500 | 97% | Read |
| Mistral | $0.150 | $0.500 | 100% | Read |
| Particle AI · nvfp4 | $0.130 | $0.500 | 100% | Read |
| Cohere · nvfp4 | $0.150 | $0.500 | 98% | Read |
| Nebius · fp4 | $0.150 | $0.500 | 100% | Read |
| Modal · fp4 | $0.150 | $0.500 | 100% | Read |
| OpenInference · fp4 | $0.060 | $0.500 | 100% | Read |
| Near AI · fp8 | $0.080 | $0.500 | 100% | Read |
| Crusoe · fp4 | $0.150 | $0.500 | 100% | Read |
| Fireworks | $0.150 | $0.500 | 100% | Read |
| Together | $0.150 | $0.500 | 100% | Read |
| Parasail · fp4 | $0.150 | $0.500 | 100% | Read |
| BaseTen · fp8 | $0.150 | $0.500 | 100% | Read |
| Venice | $0.150 | $0.500 | 100% | Read |
| DeepInfra · fp4 | $0.150 | $0.500 | 100% | Read |
| Z.AI · fp8 | $0.150 | $0.500 | 100% | Read |
| Parasail · fp4 | $0.188 | $0.625 | 100% | Read |
| Sail Research · fp4 | $0.175 | $0.625 | 100% | Read |
| Sail Research · fp4 | $0.175 | $0.625 | 100% | Read |
| Fireworks | $0.225 | $0.750 | 96% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text, image, video, pdf
- Writes: text
- Weights: Open weights
- Reasoning: Yes
- Tool calling: Yes
- Longest reply: 131K tokens
- Repeated text is billed at $0.030 per 1M (prompt caching)
Compare with
Appears in
Estimate your monthly cost with GLM 5.3 Flash
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.
