ModelPriceWatch.com
Last scan 2026-10-12 Models tracked 280 Providers 37 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

GLM-5.3-Flash

by Z.AI · 320B (A18B) parameters

Current budget open weights cheap tier
Email me when GLM-5.3-Flash gets cheaper

Today's price · per 1M tokens

Input

$0.150

per 1M tokens

Output

$0.500

per 1M tokens

Blended

$0.237

blended $/1M — 3:1 weighted input:output

Cached input

$0.030

20% of input — prompt caching

How this price is scoped: Standard list pricing, per 1M tokens. Z.AI's 50% launch discount for GLM-5.3-Flash ended at 24:00 on September 9, 2026 (UTC+8, Singapore time). These are the list prices that its own rate card (docs.z.ai/guides/overview/pricing) printed, struck through, beside the discounted ones throughout the promotion: input $0.15, cached input $0.03, output $0.50 per 1M tokens. Cached-input STORAGE is separately labelled "Limited-time Free" with no end date, and is not a rate we publish.

Source: official Z.AI pricing · read MODELPRICEWATCH.COM · 2026-10-12

Price receipt

We read Z.AI's own pricing page on and found GLM-5.3-Flash listed with a price on the same row — that read is where the number above comes from, recorded as a price increase. We keep the page text we read; its content hash is aac79925f8.

Overview

Z.AI's first natively multimodal GLM-5 model and a separate SKU from GLM-5.3, not a cheaper serving of it — 320B total / 18B activated, video, image, text and file input (images via image_url content blocks), 1M-token context. Priced at $0.15/$0.50 per 1M tokens with cached input at $0.03 — the list rate that took effect on 2026-09-10, when Z.AI's 50% launch discount ended. Weights are published under MIT at huggingface.co/zai-org/GLM-5.3-Flash.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

RunPod →
Rent GPUs by the second and serve any open-weight model yourself.
Novita AI →
Serverless open-model inference billed per token — no GPU ops.
Vast.ai →
GPU marketplace — rent spot/on-demand GPUs cheaply and serve any open-weight model.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 3/4
Text ✓ Image ✓ Audio Video ✓
Output
Text ✓
Features 4/9
Prompt caching ✓ Reasoning Coding Fast inference Long context ✓ Open weights ✓ Multimodal ✓ Web search Realtime

Benchmark performance

accuracy % · higher is better
Avg benchmark score69.8%
Perf per $/Mtok293.9
Terminal-Bench 2.1
84.3%
Humanity's Last Exam
55.3%
AutoBench *
48.8%
Independent composite scores
  • AA Intelligence Index: 41.8 (Artificial Analysis composite across reasoning, knowledge and coding evals)
  • AA-Omniscience Index: 7.5 (−100–100)
  • GDPval-AA v2: 1655 ELO (Real-world work tasks, human baseline = 1000)
  • LMSYS Chatbot Arena: 1475.4 ELO (Human preference)
How it stacks up
  • Ranks #47 of 98 benchmarked models by average score
  • Ranks #11 of 193 comparably-measured models by percentile score, across 6 independent measurements
  • Ranks #5 of 193 comparably-tested models by normalized performance per dollar top value
  • Strongest at AutoBench — 48.8%, #1 of 13

Source: Vellum LLM Leaderboard · updated · See full rankings →

Specifications

Provider
Z.AI
Context window
1M tokens
Modality
text, image, video
Parameters
320B (A18B)
Open source
Yes — open weights available
Released
Status
Current
Last updated
Tags
open-weightsbudgetmultimodalcachinglong-context

Availability verified: — listed on Z.AI's own page

Try on Z.AI Official pricing page