Gemini 3.8 Flash vs GLM 5.3

Google against Z.AI. Prices, scores and speed below are live.

  • GLM 5.3 costs 76% less on blended price ($0.60 against $2.48 per 1M tokens, 3 input to 1 output).
  • GLM 5.3 scores higher on the Artificial Analysis Intelligence Index: 31.7 against 30.8.
  • On the 38 benchmarks both have scores on, Gemini 3.8 Flash leads on 26 and GLM 5.3 on 12.
  • Both read up to 1M tokens of context.
  • Gemini 3.8 Flash is newer: listed Sep 2, 2026, against Aug 18, 2026.

Side by side

Gemini 3.8 FlashGLM 5.3
LabGoogleZ.AI
ListedSep 2, 2026Aug 18, 2026
Input per 1M tokens$1.24$0.10
Output per 1M tokens$6.19$2.10
Cached input per 1M$0.12$0.10
Blended (3:1) per 1M$2.48$0.60
Context window1M1M
Max output66K944K
Knowledge cutoff--
Intelligence index30.831.7
Coding index76.374.8
ReasoningYesYes
Reads imagesYesNo
Tool useYesYes
Structured outputYesYes

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkGemini 3.8 FlashGLM 5.3
AIME98.9%Epoch AI91.1%Epoch AI
APEX-Agents64.3%Epoch AI56.6%Epoch AI
Artificial Analysis Agentic Index40.2Artificial Analysis and OpenRouter53.1Artificial Analysis and OpenRouter
Artificial Analysis Coding Index76.3Artificial Analysis and OpenRouter74.8Artificial Analysis and OpenRouter
Artificial Analysis Intelligence Index30.8Artificial Analysis and OpenRouter31.7Artificial Analysis and OpenRouter
Chess Puzzles58%Epoch AI37%Epoch AI
DeepSWE73.8%Epoch AI69%Epoch AI
Design Arena: 3D1285 EloDesign Arena1311 EloDesign Arena
Design Arena: ASCII Art1274 EloDesign Arena1206 EloDesign Arena
Design Arena: Code1298 EloDesign Arena1291 EloDesign Arena
Design Arena: Data Viz1273 EloDesign Arena1252 EloDesign Arena
Design Arena: Game Dev1300 EloDesign Arena1295 EloDesign Arena
Design Arena: Mobile Apps (agentic)1232 EloDesign Arena1190 EloDesign Arena
Design Arena: PPTX Slides (agentic)1129 EloDesign Arena1208 EloDesign Arena
Design Arena: SVG1292 EloDesign Arena1258 EloDesign Arena
Design Arena: UI Component1308 EloDesign Arena1290 EloDesign Arena
Design Arena: Web Apps (agentic)1227 EloDesign Arena1275 EloDesign Arena
Design Arena: Website1297 EloDesign Arena1286 EloDesign Arena
DTBench95.7%Epoch AI87.7%Epoch AI
FrontierCode41.2Epoch AI40.1Epoch AI
FrontierMath Tier 4 (v2)22%Epoch AI29.3%Epoch AI
FrontierMath Tiers 1-3 (v2)68.4%Epoch AI68.8%Epoch AI
GPQA Diamond95.6%Artificial Analysis and OpenRouter85.4%Artificial Analysis and OpenRouter
LMArena (Coding)1511 EloLMArena1499 EloLMArena
LMArena (Overall)1500 EloLMArena1476 EloLMArena
LMArena: Creative Writing1494 EloLMArena1460 EloLMArena
LMArena: Hard Prompts1511 EloLMArena1493 EloLMArena
LMArena: Longer Query1509 EloLMArena1487 EloLMArena
LMArena: Math1518 EloLMArena1504 EloLMArena
LMArena: Multi-Turn1503 EloLMArena1474 EloLMArena
LMCA52.9Epoch AI55.5Epoch AI
Mystery Game Puzzles47%Epoch AI33%Epoch AI
ProofBench48%Epoch AI49%Epoch AI
SciCode56.6%Epoch AI59%Epoch AI
SimpleQA Verified69.7%Epoch AI41%Epoch AI
tau-bench (Airline)71.3%Artificial Analysis and OpenRouter77.3%Artificial Analysis and OpenRouter
WebDev Arena1568 EloEpoch AI1614 EloEpoch AI
WeirdML84.8%Epoch AI75.4%Epoch AI

In our tests and guides

Where Gemini 3.8 Flash or GLM 5.3 ran on a real brief, or shows in a live table.

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See Google models, Z.AI models, every LLM's price and our sources.