GLM 5.2 Fast
GLM 5.2 Fast is a faster-serving variant of GLM-5.2, Z.ai's flagship long-horizon coding model, released June 23, 2026. Same weights and 1M tokens context, served on inference infrastructure tuned for higher throughput.
import { streamText } from 'ai'
const result = streamText({ model: 'zai/glm-5.2-fast', prompt: 'Why is the sky blue?'})Playground
Try out GLM 5.2 Fast by Z.ai. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
GLM 5.2 Fast
Ask GLM 5.2 Fast anything to try it out.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
| Provider |
|---|
 Wafer | 1M | 2.1s | 166tps | $3.00/M | $10.25/M | Read:$0.5/M Write:— | — | — | +1 | 06/23/2026 | ||
 Fireworks | 1M | 0.7s | 220tps | $2.10/M | $6.60/M | Read:$0.21/M Write:— | — | — | +1 | 06/23/2026 |
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by Z.ai
| Model |
|---|
| 1M | 1.1s | 158tps | $0.95/M | $3.00/M | Read:$0.18/M Write:— | — | — | +1 |   +3 | 06/16/2026 | |||
| 205K | 0.4s | 106tps | $1.30/M | $4.30/M | Read:$0.26/M Write:— | — | — | +1 | +3 | 04/07/2026 | |||
| 203K | 5.7s | 32tps | $1.20/M | $4.00/M | Read:$0.24/M Write:— | — | — | +1 | | 03/15/2026 | |||
| 203K | 0.2s | 72tps | $0.80/M | $2.56/M | Read:$0.16/M Write:— | — | — | +1 |  +3 | 02/12/2026 | |||
| 200K | 1.7s | 165tps | $0.06/M | $0.40/M | Read:$0.01/M Write:— | — | — | +1 | | 01/19/2026 | |||
| 205K | 0.1s | 429tps | $2.25/M | $2.75/M | Read:$2.25/M Write:— | — | — | +1 |  +3 | 12/22/2025 |
About GLM 5.2 Fast
GLM 5.2 Fast was released June 23, 2026 as the faster-serving variant of GLM-5.2, Z.ai's flagship model for long-horizon coding and agentic engineering. GLM 5.2 Fast runs the same underlying weights, so output quality matches the standard glm-5.2 endpoint. The difference is the serving stack: providers behind GLM 5.2 Fast run inference infrastructure tuned for higher throughput, and AI Gateway publishes live throughput and latency metrics on this page.
Everything that defines GLM-5.2 carries over. You get the 1M tokens context window, thinking on or off per request, selectable reasoning effort, tool calling, structured output, streaming, and implicit caching. Agent loops benefit the most: when a task chains dozens of model calls, faster serving shortens every step, and the savings compound across the whole run.
Speed comes at a higher per-token rate than the standard endpoint, so most teams split traffic. Interactive and user-facing paths route to GLM 5.2 Fast, while batch and background jobs stay on glm-5.2. Both share the same API surface through AI Gateway, so switching is a one-line model identifier change. The AI SDK, Chat Completions API, Responses API, Messages API, and other API formats all work.
What To Consider When Choosing a Provider
- Configuration: Quality is identical to
glm-5.2because the weights are the same. Choose between the two on serving characteristics and cost, not capability. Check the live throughput and latency metrics on this page and on theglm-5.2page before committing traffic. - Configuration: Per-token rates run higher than the standard endpoint. Route interactive, latency-sensitive paths here and keep batch or overnight jobs on
glm-5.2to control spend. - Configuration: Reasoning effort still applies. A high effort setting spends more tokens and time regardless of serving speed, so tune effort per task instead of relying on fast infrastructure to absorb deep deliberation.
- Zero Data Retention: AI Gateway supports Zero Data Retention for this model via direct gateway requests (BYOK is not included). To configure this, check the documentation.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use GLM 5.2 Fast
Best For
- Interactive Coding Agents: Editor and terminal agents where users wait on every response
- Multi-Step Agent Loops: Chained tool calls where per-step speed compounds across the run
- User-Facing AI Products: Chat and assistant features that need GLM-5.2-class quality without long pauses
- Rapid Iteration Workflows: Tight edit-test-fix cycles that keep developers in flow
Consider Alternatives When
- Batch or Background Jobs:
glm-5.2delivers identical output quality at lower per-token cost - High-Volume Lightweight Tasks: GLM-5-Turbo handles extraction and classification workloads economically
- Vision or GUI Input: GLM-5V-Turbo adds screenshot and image understanding to the GLM-5 generation
- Maximum Cost Efficiency: GLM-4.7-Flash covers simple prompts when GLM-5.2-class capability is unnecessary
Conclusion
GLM 5.2 Fast removes the usual speed-versus-quality tradeoff for GLM-5.2 workloads: same weights, faster serving. Route latency-sensitive traffic here, keep batch work on glm-5.2, and let AI Gateway's live metrics and unified API make the split a one-line decision.