GLM-5.3-Flash
GLM-5.3-Flash is an open-weights language model from Z.ai, released August 20, 2026. It has 321B parameters (mixture-of-experts, 25B active). The context window is 1M tokens. On the Epoch Capabilities Index it scores 153.8, rank 52 of 256 rated models. API price: $0.15 input / $0.5 output per million tokens.
We introduce GLM-5.3-Flash, the first natively multimodal model in the GLM-5 series. With 320B total parameters and just 18B active parameters, it outperforms GLM-5.2 across benchmarks and real-world workloads at one-tenth the price, while approaching Claude Opus 4.8 on coding and agentic benchmarks.
Check it against your own PC →
Can you run GLM-5.3-Flash locally?
Even with 32 GB of VRAM and 32 GB of RAM it is too large to run at a usable speed on a single consumer GPU. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Too large | – | – | – | – |
| 12 GB VRAM | Too large | – | – | – | – |
| 16 GB VRAM | Too large | – | – | – | – |
| 24 GB VRAM | Too large | – | – | – | – |
| 32 GB VRAM | Too large | – | – | – | – |
Ready-made quantizations on Hugging Face: peasantsmith/GLM-5.3-Flash-Maya-GGUF · unsloth/GLM-5.3-Flash-GGUF · patrickbdevaney/GLM-5.3-Flash-REAP50-GGUF · GCSA-AiLab/GLM-5.3-Flash-Uncensored-RCO-GSQ-GGUF · huihui-ai/Huihui-GLM-5.3-Flash-abliterated-GGUF · pfeifferj/GLM-5.3-Flash-GSQ-RCO-GGUF
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 13.0 | Chess Puzzles, Mystery Game Puzzles |
| Math | 41.7 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Coding | 56.0 | FrontierSWE, DeepSWE |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 93.9% |
| GPQA Diamond | 90.5% |
| DeepSWE | 63.4% |
| FrontierMath (Tiers 1–3) | 58.7% |
| ProofBench | 21.0% |
| FrontierSWE | 18.1% |
| FrontierMath-Tier-4-v2-Private | 17.1% |
| Chess Puzzles | 16.9% |
| Mystery Game Puzzles | 0.0% |
Links
Announcement · Hugging Face: zai-org/GLM-5.3-Flash · Epoch AI benchmarks
Models with a similar score
Grok 4.6 · Qwen 3.8 Max · Claude Opus 4.7 · GPT-5.2 Pro · GLM-5.3 · Claude Sonnet 5 · DeepSeek V4.1 Flash · GPT-5.6 Luna
More from Z.ai
GLM-5.3-FlashX · GLM 5.3 Flash BF16 · GLM 5.3 BF16 · GLM-5.3 · GLM-5.2 · GLM-5.1 · GLM-5V-Turbo · GLM-5-Turbo