GLM 5.3 Flash vs GPT-5.3-Codex
GPT-5.3-Codex leads GLM 5.3 Flash on coding, 60.0 to 57.7 on the Vibe Coding Index, and GLM 5.3 Flash costs 28.0x less per output token ($0.50 vs $14.00 per million). Updated October 22, 2026; 22 benchmarks have a result for both.
GLM 5.3 Flash or GPT-5.3-Codex: which to pick
Pick GPT-5.3-Codex for coding: it scores 60.0 on the Vibe Coding Index against 57.7 for GLM 5.3 Flash, though on 3 benchmarks against 6.
Pick GLM 5.3 Flash for output tokens 28.0x cheaper ($0.50 vs $14.00 per million) and a lower cost per solved task ($0.02 against $0.59).
- Shared results
- 22
- GLM 5.3 Flash only
- 80
- GPT-5.3-Codex only
- 12
- Like-for-like categories
- 5 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
GPT-5.3-Codex
GPT-5.3-Codex scores 60.0 against 57.7.
6 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
GLM 5.3 Flash
GLM 5.3 Flash leads, 41.8 against 32.5.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only GLM 5.3 Flash has a AA Agentic Index result, 50.9.
Directional only
Reasoning
GPQA Diamond
No call
Only GLM 5.3 Flash has a GPQA Diamond result, 86.4%.
Directional only
Knowledge
MMLU Pro
No call
Only GLM 5.3 Flash has a MMLU Pro result, 86.1%.
Directional only
Web design
Design Arena: websites
GLM 5.3 Flash
GLM 5.3 Flash leads, 1271 against 1162.
Same board, both measured
General
Vals Index
No call
Only GLM 5.3 Flash has a Vals Index result, 50.9%.
Directional only
Long documents
Listed context window
GLM 5.3 Flash
GLM 5.3 Flash takes 1.05M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GLM 5.3 Flash
GLM 5.3 Flash $0.0004, GPT-5.3-Codex $0.0088 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GLM 5.3 Flash
GLM 5.3 Flash $0.02, GPT-5.3-Codex $0.59.
Published attempt cost and pass rate
At a glance
- Coding index
- GLM 5.3 Flash 57.7 from 6 benchmarks, GPT-5.3-Codex 60.0 from 3 benchmarks.
- Widest category
- Coding, 23.3 points apart on average over 9 shared benchmarks.
- Widest single gap
- ALE-Bench: GLM 5.3 Flash 303.6 against GPT-5.3-Codex 1655.2, 45.1 points apart on a 0 to 100 scale.
- Output price
- GLM 5.3 Flash $0.50 against GPT-5.3-Codex $14.00 per million output tokens, 28.0x apart.
- Context window
- GLM 5.3 Flash takes 1.05M tokens, GPT-5.3-Codex 400K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GLM 5.3 Flash | Bars | GPT-5.3-Codex | Ties |
|---|---|---|---|---|
| Coding | 5 | 4 | ||
| Intelligence | 1 | 1 | ||
| Reasoning | 0 | 2 | ||
| Web design | 8 | 0 | ||
| General | 0 | 1 | ||
| All shared boards | 14 | 8 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index54
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| LiveBench coding | 80.6% | Not measured | Only GLM 5.3 Flash measured |
| LMArena WebDev | 1607 | 1407 | GLM 5.3 Flash |
| AA Coding Index | 71.5 | Not measured | Only GLM 5.3 Flash measured |
| LiveCodeBench (Vals AI) | 80.5% | 87.3% | GPT-5.3-Codex |
| SWE-bench Verified (Vals AI) | 92.0% | 78.0% | GLM 5.3 Flash |
| Terminal-Bench 4 (Vals AI) | 37.4% | Not measured | Only GLM 5.3 Flash measured |
| Vibe Code Bench (Vals AI) | 30.8% | 61.8% | GPT-5.3-Codex |
| Code migration (Vals AI) | 20.5% | Not measured | Only GLM 5.3 Flash measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only GLM 5.3 Flash measured |
| IOI (Vals AI) | 52.5% | 53.8% | GPT-5.3-Codex |
| Terminal-Bench 2.1 (Vals AI) | 62.9% | Not measured | Only GLM 5.3 Flash measured |
| Vibe Code Bench 1-100 (Vals AI) | 16.0% | Not measured | Only GLM 5.3 Flash measured |
| SkillsBench (Vals AI) | 40.2% | Not measured | Only GLM 5.3 Flash measured |
| CyberBench (Vals AI) | 67.9% | Not measured | Only GLM 5.3 Flash measured |
| SciCode | 51.6% | Not measured | Only GLM 5.3 Flash measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 53.0% | Only GPT-5.3-Codex measured |
| Terminal-Bench | Not measured | 78.4% | Only GPT-5.3-Codex measured |
| FrontierCode | 31.8% | Not measured | Only GLM 5.3 Flash measured |
| FrontierSWE | 18.1% | Not measured | Only GLM 5.3 Flash measured |
| WeirdML | Not measured | 79.3% | Only GPT-5.3-Codex measured |
| ALE-Bench | 303.6 | 1655.2 | GPT-5.3-Codex |
| LMArena Text: coding | 1507 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: software and IT | 1500 | Not measured | Only GLM 5.3 Flash measured |
| LMArena WebDev: React | 1624 | 1395 | GLM 5.3 Flash |
| LMArena WebDev: HTML | 1577 | 1422 | GLM 5.3 Flash |
| LMArena WebDev: from an image | 1578 | 1443 | GLM 5.3 Flash |
| Neither model has a result on the other 17 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| AA Intelligence Index | 41.8 | 32.5 | GLM 5.3 Flash |
| Epoch Capabilities Index | 153.1 | 157.1 | GPT-5.3-Codex |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index20 boards, none shared
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| AA Agentic Index | 50.9 | Not measured | Only GLM 5.3 Flash measured |
| Design Arena: agents building web apps | Not measured | 1032 | Only GPT-5.3-Codex measured |
| Design Arena: agents building full-stack apps | Not measured | 983 | Only GPT-5.3-Codex measured |
| Design Arena: agents building mobile apps | Not measured | 1053 | Only GPT-5.3-Codex measured |
| Design Arena: agents building Android apps | Not measured | 1084 | Only GPT-5.3-Codex measured |
| Design Arena: agents building Godot games | Not measured | 1124 | Only GPT-5.3-Codex measured |
| LiveBench research | 70.1% | Not measured | Only GLM 5.3 Flash measured |
| LiveBench browser | 84.0% | Not measured | Only GLM 5.3 Flash measured |
| LiveBench long agentic | 28.4% | Not measured | Only GLM 5.3 Flash measured |
| SREBench (Vals AI) | 0.8% | Not measured | Only GLM 5.3 Flash measured |
| Finance Agent (Vals AI) | 57.9% | Not measured | Only GLM 5.3 Flash measured |
| Tax Agent Bench (Vals AI) | 31.6% | Not measured | Only GLM 5.3 Flash measured |
| Legal Research Bench (Vals AI) | 45.2% | Not measured | Only GLM 5.3 Flash measured |
| Harvey's Legal Agent Benchmark (Vals AI) | 10.4% | Not measured | Only GLM 5.3 Flash measured |
| Excel Modeling Benchmark (Vals AI) | 55.9% | Not measured | Only GLM 5.3 Flash measured |
| Terminal-Bench-Science (Vals AI) | 2.9% | Not measured | Only GLM 5.3 Flash measured |
| τ²-Bench Telecom | Not measured | 86.0% | Only GPT-5.3-Codex measured |
| APEX-Agents | 52.8% | Not measured | Only GLM 5.3 Flash measured |
| METR time horizons | Not measured | 74.5% | Only GPT-5.3-Codex measured |
| Vending-Bench 2 | Not measured | $5,940 | Only GPT-5.3-Codex measured |
| Neither model has a result on the other 22 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| GPQA Diamond | 86.4% | Not measured | Only GLM 5.3 Flash measured |
| LiveBench reasoning | 77.6% | Not measured | Only GLM 5.3 Flash measured |
| MMMU Pro (Vals AI) | 86.0% | Not measured | Only GLM 5.3 Flash measured |
| GPQA Diamond (Artificial Analysis run) | 91.2% | 91.5% | GPT-5.3-Codex |
| Humanity's Last Exam (Artificial Analysis run) | 39.9% | 42.5% | GPT-5.3-Codex |
| GPQA Diamond (Epoch AI run) | 90.2% | Not measured | Only GLM 5.3 Flash measured |
| ARC-AGI | 91.0% | Not measured | Only GLM 5.3 Flash measured |
| ARC-AGI-2 | 65.8% | Not measured | Only GLM 5.3 Flash measured |
| Chess puzzles | 21.0% | Not measured | Only GLM 5.3 Flash measured |
| Mystery game puzzles | 8.0% | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: hard prompts | 1490 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: expert | 1511 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: hard prompts in English | 1488 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: diagrams | 1317 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: homework | 1318 | Not measured | Only GLM 5.3 Flash measured |
| Neither model has a result on the other 8 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| OTIS Mock AIME | 93.9% | Not measured | Only GLM 5.3 Flash measured |
| FrontierMath, tiers 1 to 3 | 55.8% | Not measured | Only GLM 5.3 Flash measured |
| FrontierMath, tier 4 | 17.1% | Not measured | Only GLM 5.3 Flash measured |
| ProofBench | 21.0% | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: math | 1503 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: mathematical work | 1497 | Not measured | Only GLM 5.3 Flash measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| MMLU Pro | 86.1% | Not measured | Only GLM 5.3 Flash measured |
| LegalBench (Vals AI) | 83.9% | Not measured | Only GLM 5.3 Flash measured |
| TaxEval (Vals AI) | 75.6% | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: science | 1484 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: medicine and healthcare | 1475 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: legal and government | 1473 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: business and finance | 1460 | Not measured | Only GLM 5.3 Flash measured |
| Neither model has a result on the other 13 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| Design Arena: websites | 1271 | 1162 | GLM 5.3 Flash |
| Design Arena: UI components | 1314 | 1135 | GLM 5.3 Flash |
| Design Arena: game dev | 1277 | 1161 | GLM 5.3 Flash |
| Design Arena: data viz | 1262 | 1164 | GLM 5.3 Flash |
| Design Arena: 3D | 1315 | 1007 | GLM 5.3 Flash |
| Design Arena: all code categories | 1281 | 1144 | GLM 5.3 Flash |
| Design Arena: SVG | 1284 | 1128 | GLM 5.3 Flash |
| Design Arena: ASCII art | 1267 | 1151 | GLM 5.3 Flash |
| LiveBench design | 77.2% | Not measured | Only GLM 5.3 Flash measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| Vals Index | 50.9% | Not measured | Only GLM 5.3 Flash measured |
| LiveBench data analysis | 76.4% | Not measured | Only GLM 5.3 Flash measured |
| MedScribe (Vals AI) | 88.9% | Not measured | Only GLM 5.3 Flash measured |
| IFBench | Not measured | 75.4% | Only GPT-5.3-Codex measured |
| Long context reasoning (AA-LCR) | 80.0% | 83.3% | GPT-5.3-Codex |
| Surface Evolver Bench | 52.5% | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text | 1471 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: instruction following | 1478 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: creative writing | 1441 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision | 1294 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: English | 1474 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: non-English | 1462 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Chinese | 1531 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: German | 1462 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Japanese | 1444 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Korean | 1436 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: French | 1491 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Spanish | 1475 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Russian | 1465 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: Polish | 1450 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: multi-turn | 1469 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: longer queries | 1481 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: writing and language | 1455 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Text: entertainment and media | 1440 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: English | 1295 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: Chinese | 1363 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: reading text | 1299 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: humor | 1279 | Not measured | Only GLM 5.3 Flash measured |
| LMArena Vision: creative writing | 1270 | Not measured | Only GLM 5.3 Flash measured |
| Neither model has a result on the other 19 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GLM 5.3 Flash | GPT-5.3-Codex | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GLM 5.3 Flash costs less
Repository review
50K input + 3K output tokens
GLM 5.3 Flash costs less
Agent loop
200K input + 10K output tokens
GLM 5.3 Flash costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Z.ai GLM 5.3 Flash | OpenAI GPT-5.3-Codex |
|---|---|---|
| Maker | Z.ai | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | August 26, 2026 | February 24, 2026 |
| Context window | 1.05M | 400K |
| Input $/Mtok | $0.15 | $1.75 |
| Output $/Mtok | $0.50 | $14.00 |
| Coding index | 57.7 | 60.0 |
| Independent measurements behind it | 6 | 3 |
| Cost per solved task | $0.02 | $0.59 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GLM 5.3 Flash better than GPT-5.3-Codex for coding?
Pick GPT-5.3-Codex for coding: it scores 60.0 on the Vibe Coding Index against 57.7 for GLM 5.3 Flash, though on 3 benchmarks against 6. Pick GLM 5.3 Flash for output tokens 28.0x cheaper ($0.50 vs $14.00 per million) and a lower cost per solved task ($0.02 against $0.59).
Which is cheaper, GLM 5.3 Flash or GPT-5.3-Codex?
GLM 5.3 Flash costs $0.15 per million input tokens and $0.50 per million output tokens; GPT-5.3-Codex costs $1.75 and $14.00. Per solved SWE-bench Verified task, GLM 5.3 Flash costs $0.02 and GPT-5.3-Codex $0.59.
Which has the bigger context window, GLM 5.3 Flash or GPT-5.3-Codex?
GLM 5.3 Flash, with 1.05M tokens against 400K.
How many benchmarks is the GLM 5.3 Flash vs GPT-5.3-Codex comparison based on?
22 benchmarks have a result for both models as of October 22, 2026. GLM 5.3 Flash's coding index rests on 6 independent benchmarks and GPT-5.3-Codex's on 3. Where only one model has a result, the page says so instead of naming a winner.