GLM-5.3-FlashReady
glm-5-3-flash
Serverless APILLMReasoningVisionPrompt cachingOpen weights
The low-latency tier of GLM-5.3, built for high-volume production traffic. It ranks #16 of the top 20 frontier models on the Artificial Analysis Intelligence Index with a score of 41.8. Released August 26, 2026.
1M
Context
320B
Parameters
#16
Intelligence rank · 41.8
Open
Weights · MIT
View full spec
ReleasedAug 26, 2026
Knowledge cutoff—
Size classlarge
Input modalitiestext, image
Output modalitiestext
Median output speed71 tok/s
Median time to first token2.8s
LicenseMIT
Usage guide
1. Choose how you'd like to use GLM-5.3-Flash
☁️ Serverless
Use with pay-per-token pricing. Zero setup, no cold starts, automatic failover between upstream routes.
Usage tiers · Standard • Priority
Try it — sign in
curl https://tokenhub-staging.coality.io/v1/chat/completions \
-H "Authorization: Bearer $TOKEN_HUB_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-5-3-flash",
"messages": [{"role": "user", "content": "Explain vector databases in one paragraph"}]
}'Pricing
| Bucket | Price | Notes |
|---|---|---|
| Input tokens | $0.1875 | per 1M tokens |
| Output tokens | $0.6250 | per 1M tokens |
| Cache read | $0.0375 | cached input · 80% off |
| Cache write | — | per 1M tokens |
Billed per token with micro-dollar precision · effective Oct 26, 2026 · default key limits 600 req/min · 200K tokens/min
Benchmarks
GPQA Diamond
91.2
Humanity's Last Exam
39.9
SciCode
51.6
Terminal-Bench 4.0
32.8
Long Context Reasoning
80.0
Frontier Code
41.0
Scores from the Artificial Analysis evaluation suite (0–100 scale).
About this model
320B-parameter open-weight model, with a 1M context window, native image input, visible reasoning traces, prompt caching at a 80% discount on cached input, 71 output tokens/s median serving speed and 2.8s to first token.