DeepSeek V4 Flash 0731 vs GLM 5.3
Updated October 21, 2026. 50 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731.
Pick DeepSeek V4 Flash 0731 when output tokens cost 11.1x less, or when a solved task costs $0.01 against $0.36.
- Shared results
- 50
- DeepSeek V4 Flash 0731 only
- 5
- GLM 5.3 only
- 50
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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
GLM 5.3
GLM 5.3 scores 70.6 against 51.8.
4 and 6 benchmarks behind the scores
Intelligence
AA Intelligence Index
GLM 5.3
GLM 5.3 leads, 44.8 against 34.3.
Same board, both measured
Agentic work
AA Agentic Index
GLM 5.3
GLM 5.3 leads, 53.1 against 41.0.
Same board, both measured
Reasoning
GPQA Diamond
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 leads, 89.9% against 88.1%.
Same board, both measured
Knowledge
MMLU Pro
GLM 5.3
GLM 5.3 leads, 86.8% against 86.2%.
Same board, both measured
Web design
Design Arena: websites
GLM 5.3
GLM 5.3 leads, 1295 against 1238.
Same board, both measured
General
Vals Index
No call
Only GLM 5.3 has a Vals Index result, 59.7%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.0003, GLM 5.3 $0.0036 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 $0.01, GLM 5.3 $0.36.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, GLM 5.3 70.6 from 6 benchmarks.
- Widest category
- Web design, 11.7 points apart on average over 8 shared benchmarks.
- Widest single gap
- IOI (Vals AI): DeepSeek V4 Flash 0731 32.7% against GLM 5.3 68.4%, 35.7 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Flash 0731 $0.396 against GLM 5.3 $4.40 per million output tokens, 11.1x apart.
- Context window
- Both take 1.05M tokens.
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 | DeepSeek V4 Flash 0731 | Bars | GLM 5.3 | Ties |
|---|---|---|---|---|
| Coding | 2 | 12 | 1 tie | |
| Intelligence | 0 | 2 | ||
| Agentic work | 1 | 5 | ||
| Reasoning | 3 | 3 | ||
| Math | 2 | 2 | ||
| Knowledge | 0 | 5 | ||
| Web design | 0 | 8 | ||
| General | 1 | 2 | 1 tie | |
| All shared boards | 9 | 39 | 2 ties |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index212
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 82.8% | Only GLM 5.3 measured |
| LMArena WebDev | Not measured | 1621 | Only GLM 5.3 measured |
| AA Coding Index | 69.1 | 74.8 | GLM 5.3 |
| LiveCodeBench (Vals AI) | 87.3% | 80.5% | DeepSeek V4 Flash 0731 |
| SWE-bench Verified (Vals AI) | 88.8% | 95.4% | GLM 5.3 |
| Terminal-Bench 4 (Vals AI) | 18.7% | 38.9% | GLM 5.3 |
| Vibe Code Bench (Vals AI) | 74.7% | 78.1% | GLM 5.3 |
| Code migration (Vals AI) | 38.6% | 44.2% | GLM 5.3 |
| ProgramBench (Vals AI) | 0.0% | 1.5% | GLM 5.3 |
| IOI (Vals AI) | 32.7% | 68.4% | GLM 5.3 |
| Terminal-Bench 2.1 (Vals AI) | 67.0% | 71.5% | GLM 5.3 |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 20.0% | Only GLM 5.3 measured |
| SkillsBench (Vals AI) | 50.7% | 47.5% | DeepSeek V4 Flash 0731 |
| CyberBench (Vals AI) | 72.1% | 72.1% | Tie |
| SciCode | 50.3% | 59.0% | GLM 5.3 |
| FrontierCode | 18.8% | 40.1% | GLM 5.3 |
| FrontierSWE | Not measured | 30.2% | Only GLM 5.3 measured |
| WeirdML | 63.0% | 75.4% | GLM 5.3 |
| ALE-Bench | 1306.1 | 1317.4 | GLM 5.3 |
| LMArena Text: coding | Not measured | 1497 | Only GLM 5.3 measured |
| LMArena Text: software and IT | Not measured | 1493 | Only GLM 5.3 measured |
| LMArena WebDev: React | Not measured | 1641 | Only GLM 5.3 measured |
| LMArena WebDev: HTML | Not measured | 1595 | Only GLM 5.3 measured |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 34.3 | 44.8 | GLM 5.3 |
| Epoch Capabilities Index | 154.8 | 156.0 | GLM 5.3 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index15
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| AA Agentic Index | 41.0 | 53.1 | GLM 5.3 |
| Design Arena: agents building mobile apps | Not measured | 1194 | Only GLM 5.3 measured |
| Design Arena: HTML slides | Not measured | 1189 | Only GLM 5.3 measured |
| Design Arena: PowerPoint slides | Not measured | 1255 | Only GLM 5.3 measured |
| LiveBench research | Not measured | 73.2% | Only GLM 5.3 measured |
| LiveBench browser | Not measured | 0.0% | Only GLM 5.3 measured |
| LiveBench long agentic | Not measured | 35.9% | Only GLM 5.3 measured |
| SREBench (Vals AI) | Not measured | 9.2% | Only GLM 5.3 measured |
| Finance Agent (Vals AI) | 49.5% | 55.8% | GLM 5.3 |
| Tax Agent Bench (Vals AI) | 28.1% | 39.7% | GLM 5.3 |
| Legal Research Bench (Vals AI) | 30.3% | 49.0% | GLM 5.3 |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 14.6% | Only GLM 5.3 measured |
| Excel Modeling Benchmark (Vals AI) | 57.0% | 56.3% | DeepSeek V4 Flash 0731 |
| BioMysteryBench (Vals AI) | 64.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Terminal-Bench-Science (Vals AI) | 4.3% | 5.7% | GLM 5.3 |
| APEX-Agents | Not measured | 56.6% | Only GLM 5.3 measured |
| Vending-Bench 2 | Not measured | $8,164 | Only GLM 5.3 measured |
| Neither model has a result on the other 22 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index33
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| GPQA Diamond | 89.9% | 88.1% | DeepSeek V4 Flash 0731 |
| LiveBench reasoning | Not measured | 85.8% | Only GLM 5.3 measured |
| MysteryMechanism (Vals AI) | Not measured | 23.0% | Only GLM 5.3 measured |
| GPQA Diamond (Artificial Analysis run) | 90.8% | 91.7% | GLM 5.3 |
| Humanity's Last Exam (Artificial Analysis run) | 38.6% | 42.3% | GLM 5.3 |
| GPQA Diamond (Epoch AI run) | 91.0% | 90.9% | DeepSeek V4 Flash 0731 |
| ARC-AGI | 89.0% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| ARC-AGI-2 | 61.4% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| SimpleBench | 61.1% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| Chess puzzles | 32.0% | 37.0% | GLM 5.3 |
| Mystery game puzzles | 34.0% | 33.0% | DeepSeek V4 Flash 0731 |
| LMArena Text: hard prompts | Not measured | 1490 | Only GLM 5.3 measured |
| LMArena Text: expert | Not measured | 1514 | Only GLM 5.3 measured |
| LMArena Text: hard prompts in English | Not measured | 1496 | Only GLM 5.3 measured |
| Neither model has a result on the other 7 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 94.4% | 91.1% | DeepSeek V4 Flash 0731 |
| FrontierMath, tiers 1 to 3 | 57.5% | 68.8% | GLM 5.3 |
| FrontierMath, tier 4 | 24.4% | 29.3% | GLM 5.3 |
| ProofBench | 56.0% | 49.0% | DeepSeek V4 Flash 0731 |
| LMArena Text: math | Not measured | 1495 | Only GLM 5.3 measured |
| LMArena Text: mathematical work | Not measured | 1497 | Only GLM 5.3 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 86.8% | GLM 5.3 |
| LegalBench (Vals AI) | 77.7% | 84.8% | GLM 5.3 |
| CorpFin (Vals AI) | 61.9% | Not measured | Only DeepSeek V4 Flash 0731 measured |
| TaxEval (Vals AI) | 70.7% | 72.4% | GLM 5.3 |
| MedCode (Vals AI) | 41.4% | 42.9% | GLM 5.3 |
| SimpleQA Verified | 33.6% | 41.0% | GLM 5.3 |
| LMArena Text: science | Not measured | 1498 | Only GLM 5.3 measured |
| LMArena Text: medicine and healthcare | Not measured | 1481 | Only GLM 5.3 measured |
| LMArena Text: legal and government | Not measured | 1473 | Only GLM 5.3 measured |
| LMArena Text: business and finance | Not measured | 1459 | Only GLM 5.3 measured |
| Neither model has a result on the other 7 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1238 | 1295 | GLM 5.3 |
| Design Arena: UI components | 1235 | 1314 | GLM 5.3 |
| Design Arena: game dev | 1208 | 1311 | GLM 5.3 |
| Design Arena: data viz | 1180 | 1259 | GLM 5.3 |
| Design Arena: 3D | 1209 | 1332 | GLM 5.3 |
| Design Arena: all code categories | 1230 | 1304 | GLM 5.3 |
| Design Arena: SVG | 1185 | 1293 | GLM 5.3 |
| Design Arena: ASCII art | 1090 | 1213 | GLM 5.3 |
| LiveBench design | Not measured | 77.3% | Only GLM 5.3 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | Leader |
|---|---|---|---|
| Vals Index | Not measured | 59.7% | Only GLM 5.3 measured |
| LiveBench data analysis | Not measured | 70.2% | Only GLM 5.3 measured |
| Public Benefits Bench (Vals AI) | Not measured | 68.5% | Only GLM 5.3 measured |
| MedScribe (Vals AI) | 80.4% | 88.8% | GLM 5.3 |
| Long context reasoning (AA-LCR) | 79.7% | 79.7% | Tie |
| DTBench | 90.9% | 87.7% | DeepSeek V4 Flash 0731 |
| LMCA | 41.7% | 55.5% | GLM 5.3 |
| LMArena Text | Not measured | 1474 | Only GLM 5.3 measured |
| LMArena Text: instruction following | Not measured | 1479 | Only GLM 5.3 measured |
| LMArena Text: creative writing | Not measured | 1458 | Only GLM 5.3 measured |
| LMArena Text: English | Not measured | 1482 | Only GLM 5.3 measured |
| LMArena Text: non-English | Not measured | 1459 | Only GLM 5.3 measured |
| LMArena Text: Chinese | Not measured | 1532 | Only GLM 5.3 measured |
| LMArena Text: German | Not measured | 1488 | Only GLM 5.3 measured |
| LMArena Text: Japanese | Not measured | 1460 | Only GLM 5.3 measured |
| LMArena Text: Korean | Not measured | 1460 | Only GLM 5.3 measured |
| LMArena Text: French | Not measured | 1503 | Only GLM 5.3 measured |
| LMArena Text: Spanish | Not measured | 1462 | Only GLM 5.3 measured |
| LMArena Text: Russian | Not measured | 1469 | Only GLM 5.3 measured |
| LMArena Text: Polish | Not measured | 1466 | Only GLM 5.3 measured |
| LMArena Text: multi-turn | Not measured | 1472 | Only GLM 5.3 measured |
| LMArena Text: longer queries | Not measured | 1483 | Only GLM 5.3 measured |
| LMArena Text: writing and language | Not measured | 1469 | Only GLM 5.3 measured |
| LMArena Text: entertainment and media | Not measured | 1447 | Only GLM 5.3 measured |
| Neither model has a result on the other 22 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Flash 0731 | GLM 5.3 | 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
DeepSeek V4 Flash 0731 costs less
Repository review
50K input + 3K output tokens
DeepSeek V4 Flash 0731 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V4 Flash 0731 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 | DeepSeek V4 Flash 0731 | Z.ai GLM 5.3 |
|---|---|---|
| Maker | DeepSeek | Z.ai |
| Weights | Open weights | Open weights |
| Released | July 31, 2026 | August 18, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.14 | $1.40 |
| Output $/Mtok | $0.396 | $4.40 |
| Coding index | 51.8 | 70.6 |
| Independent measurements behind it | 4 | 6 |
| Cost per solved task | $0.01 | $0.36 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Flash 0731 better than GLM 5.3 for coding?
Pick GLM 5.3 for coding: it scores 70.6 on the Vibe Coding Index against 51.8 for DeepSeek V4 Flash 0731. Pick DeepSeek V4 Flash 0731 when output tokens cost 11.1x less, or when a solved task costs $0.01 against $0.36.
Which is cheaper, DeepSeek V4 Flash 0731 or GLM 5.3?
DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.396 per million output tokens; GLM 5.3 costs $1.40 and $4.40. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and GLM 5.3 $0.36.
Which has the bigger context window, DeepSeek V4 Flash 0731 or GLM 5.3?
Neither: both take 1.05M tokens.
How many benchmarks is the DeepSeek V4 Flash 0731 vs GLM 5.3 comparison based on?
50 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Flash 0731's coding index rests on 4 independent benchmarks and GLM 5.3's on 6. Where only one model has a result, the page says so instead of naming a winner.