GLM 5.3 Flash BF16
zai-org/GLM-5.3-Flash-BF16
Open Source · chat · open-weights
Open
Alert me on changes
Intelligence
Context
1.3M
Max output
131.1K
Weights
Open
API $/1M
$0.15 / $0.50
Modalities
text · image
Released
25 Aug 2026
Intelligence Index via Artificial Analysis · 0–100, higher is better
License: mit · zai-org/GLM-5.3-Flash-BF16
AI summary
● machine-written
Z.ai releases GLM-5.3-Flash, sparse MoE model with 1M context window
GLM-5.3-Flash is a native multimodal model from Z.ai featuring sparse mixture-of-experts architecture with 320B total parameters and 18B active per token. The model supports text and image input, includes always-on reasoning capabilities, and is priced at $0.15 per million input tokens and $0.50 per million output tokens. It is suited for efficient coding and long-horizon agent tasks with a 1M-token context window.
What's new
- Released August 26, 2026 under MIT license on Hugging Face
- Achieves 63.4 Pass@1 on DeepSWE v1.1, matching independently verified benchmarks
- Hybrid sparse and linear attention architecture reduces compute overhead
- Promotional pricing at 50% discount through September 9, 2026
- Available across 20+ providers via OpenRouter with load balancing
Best for
Efficient coding and code generation tasksLong-horizon agent tasks with tool callingAgentic workflows requiring automation and reasoningLong-context applications with 1M token window