G
GLM-5.3 Flash
A 321B/18B MIT-licensed light flagship with a LiveBench design score above Gemini 4 Argon
Intelligence v5
— v4.3.2 41.8
Leaderboard rank
—
Input / output
$0.20/$0.80
Blended 3:1
$0.35
Output speed
120 tok/s
Value
—
Capability profile Normalised to the field maximum
Specifications
Vendor
Z.ai
Released
2026-09
Weights
Open (MIT)
Context
1,000,000 tokens
Max output
Undisclosed
Parameters
321B total / 18B active
Input price
$0.20 / 1M
Output price
$0.80 / 1M
Cache hit
n/a
Output speed
120 tok/s
Intelligence
41.8（v4.3.2）
Benchmark scores
How it is measured →Bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 | — | — | not indexed |
| AA Intelligence Index v4.3.2 Intelligence index (legacy) | 41.8 | 73% | −15.8 pts |
| GPQA Diamond | — | — | not indexed |
| Terminal-Bench 4.0 | — | — | not indexed |
| LiveBench Design Document generation (DOCX/PPT/PDF) | 77.2 | 89% | −9.7 pts |
| Vals Index | — | — | not indexed |
Assessment
LiveBench design 77.2, above Gemini 4 Argon's 73.7, at roughly one fifth of the price. The strongest all-round option under permissive Apache/MIT-style licences — a good fit for self-hosting.
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| MiniMax M3.1 FlashMiniMax | 29.0 | $0.26 | 118 | — |
| Claude Haiku 5.5Anthropic | 36.2 | $0.20 | 235 | — |
| DeepSeek V4.1 Flash Open weightsDeepSeek | 32.5 | $0.53 | 148 | — |
| MiMo-V2.6 Pro Open weightsXiaomi | 30.0 | $0.61 | 81 | — |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard