GLM 5.2 vs GPT-6.2 Sol
GPT-6.2 Sol leads GLM 5.2 on coding, 83.1 to 61.1 on the Vibe Coding Index, and GLM 5.2 costs 2.3x less per output token ($4.40 vs $10.00 per million). Updated November 22, 2026; 35 benchmarks have a result for both.
GLM 5.2 or GPT-6.2 Sol: which to pick
Pick GPT-6.2 Sol for coding: it scores 83.1 on the Vibe Coding Index against 61.1 for GLM 5.2.
Pick GLM 5.2 for output tokens 2.3x cheaper ($4.40 vs $10.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
- Shared results
- 35
- GLM 5.2 only
- 77
- GPT-6.2 Sol only
- 21
- Like-for-like categories
- 7 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-6.2 Sol
GPT-6.2 Sol scores 83.1 against 61.1.
4 benchmarks behind each score
Intelligence
AA Intelligence Index
No call
Only GPT-6.2 Sol has a AA Intelligence Index result, 57.3.
Directional only
Reasoning
GPQA Diamond
No call
Only GLM 5.2 has a GPQA Diamond result, 85.6%.
Directional only
Knowledge
MMLU Pro
No call
Only GLM 5.2 has a MMLU Pro result, 86.7%.
Directional only
Web design
Design Arena: websites
No call
Only GLM 5.2 has a Design Arena: websites result, 1272.
Directional only
General
Vals Index
No call
Only GPT-6.2 Sol has a Vals Index result, 66.1%.
Directional only
Long documents
Listed context window
GPT-6.2 Sol
GPT-6.2 Sol takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GLM 5.2
GLM 5.2 $0.0036, GPT-6.2 Sol $0.0070 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GLM 5.2 has the number, $0.86.
Directional only
At a glance
- Coding index
- GLM 5.2 61.1 from 4 benchmarks, GPT-6.2 Sol 83.1 from 4 benchmarks.
- Widest category
- Math, 70.7 points apart on average over 1 shared benchmark.
- Widest single gap
- MathArena: GLM 5.2 20.4% against GPT-6.2 Sol 91.2%, 70.7 points apart on a 0 to 100 scale.
- Output price
- GLM 5.2 $4.40 against GPT-6.2 Sol $10.00 per million output tokens, 2.3x apart.
- Context window
- GLM 5.2 takes 1.05M tokens, GPT-6.2 Sol 1.05M.
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 5.2 | Bars | GPT-6.2 Sol | Ties |
|---|---|---|---|---|
| Coding | 2 | 6 | ||
| Agentic work | 0 | 2 | ||
| Reasoning | 2 | 5 | ||
| Math | 0 | 1 | ||
| Knowledge | 2 | 1 | ||
| Web design | 0 | 3 | ||
| General | 11 | 0 | ||
| All shared boards | 17 | 18 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index26
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 83.7% | Only GPT-6.2 Sol measured |
| LMArena WebDev | 1608 | 1754 | GPT-6.2 Sol |
| LiveCodeBench (Vals AI) | 69.5% | Not measured | Only GLM 5.2 measured |
| SWE-bench Verified (Vals AI) | 82.8% | Not measured | Only GLM 5.2 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 60.6% | Only GPT-6.2 Sol measured |
| Vibe Code Bench (Vals AI) | 64.0% | 87.6% | GPT-6.2 Sol |
| Code migration (Vals AI) | 37.9% | Not measured | Only GLM 5.2 measured |
| ProgramBench (Vals AI) | 0.5% | Not measured | Only GLM 5.2 measured |
| Terminal-Bench 2.1 (Vals AI) | 67.8% | Not measured | Only GLM 5.2 measured |
| SkillsBench (Vals AI) | 45.1% | Not measured | Only GLM 5.2 measured |
| SciCode | 51.2% | Not measured | Only GLM 5.2 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 50.8% | Not measured | Only GLM 5.2 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 1.0% | 60.1% | GPT-6.2 Sol |
| FrontierCode 2.0 | Not measured | 44.4% | Only GPT-6.2 Sol measured |
| FrontierCode | 24.5% | Not measured | Only GLM 5.2 measured |
| WeirdML | 70.1% | Not measured | Only GLM 5.2 measured |
| MirrorCode | 36.7% | Not measured | Only GLM 5.2 measured |
| ALE-Bench | 1047.0 | Not measured | Only GLM 5.2 measured |
| LMArena Text: coding | 1488 | 1440 | GLM 5.2 |
| LMArena Text: software and IT | 1487 | 1434 | GLM 5.2 |
| LMArena WebDev: React | 1621 | 1761 | GPT-6.2 Sol |
| LMArena WebDev: HTML | 1567 | 1840 | GPT-6.2 Sol |
| SWE Atlas: Codebase QnA | 48.1% | Not measured | Only GLM 5.2 measured |
| SWE Atlas: Refactoring | 42.4% | Not measured | Only GLM 5.2 measured |
| SWE Atlas: Test Writing | 41.5% | Not measured | Only GLM 5.2 measured |
| GameBench 2 | 49.9 | 94.4 | GPT-6.2 Sol |
| Terminal-Bench 3.0 | 4.6% | Not measured | Only GLM 5.2 measured |
| ProgramBench | 0.0% | Not measured | Only GLM 5.2 measured |
| Neither model has a result on the other 24 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 57.3 | Only GPT-6.2 Sol measured |
| Epoch Capabilities Index | 153.1 | Not measured | Only GLM 5.2 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1191 | Not measured | Only GLM 5.2 measured |
| Design Arena: agents building full-stack apps | 1218 | Not measured | Only GLM 5.2 measured |
| Design Arena: agents building mobile apps | 1154 | Not measured | Only GLM 5.2 measured |
| Design Arena: agents building Android apps | 1168 | Not measured | Only GLM 5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GLM 5.2 measured |
| Design Arena: agents building games | 1156 | Not measured | Only GLM 5.2 measured |
| Design Arena: HTML slides | 1075 | Not measured | Only GLM 5.2 measured |
| Design Arena: PowerPoint slides | 1160 | Not measured | Only GLM 5.2 measured |
| Design Arena: spreadsheets | 1027 | Not measured | Only GLM 5.2 measured |
| LiveBench research | Not measured | 90.1% | Only GPT-6.2 Sol measured |
| LiveBench browser | Not measured | 88.2% | Only GPT-6.2 Sol measured |
| LiveBench long agentic | Not measured | 63.6% | Only GPT-6.2 Sol measured |
| SREBench (Vals AI) | 0.0% | Not measured | Only GLM 5.2 measured |
| Finance Agent (Vals AI) | 49.7% | Not measured | Only GLM 5.2 measured |
| Tax Agent Bench (Vals AI) | Not measured | 24.8% | Only GPT-6.2 Sol measured |
| Legal Research Bench (Vals AI) | 31.3% | 37.0% | GPT-6.2 Sol |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 12.1% | Only GPT-6.2 Sol measured |
| Excel Modeling Benchmark (Vals AI) | 61.5% | Not measured | Only GLM 5.2 measured |
| BioMysteryBench (Vals AI) | Not measured | 77.8% | Only GPT-6.2 Sol measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 58.6% | Only GPT-6.2 Sol measured |
| Vals Science Index | Not measured | 57.8% | Only GPT-6.2 Sol measured |
| τ²-Bench Telecom | 99.1% | Not measured | Only GLM 5.2 measured |
| APEX-Agents | 45.2% | 66.6% | GPT-6.2 Sol |
| PostTrainBench | 30.4% | Not measured | Only GLM 5.2 measured |
| Vending-Bench 2 | $8,314 | Not measured | Only GLM 5.2 measured |
| LMArena Search | Not measured | 1179 | Only GPT-6.2 Sol measured |
| MCP Atlas | 77.8% | Not measured | Only GLM 5.2 measured |
| Toolathlon-Verified | 59.9% | Not measured | Only GLM 5.2 measured |
| τ³-Banking | 37.1% | Not measured | Only GLM 5.2 measured |
| Neither model has a result on the other 15 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index25
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| GPQA Diamond | 85.6% | Not measured | Only GLM 5.2 measured |
| LiveBench reasoning | Not measured | 90.6% | Only GPT-6.2 Sol measured |
| MysteryMechanism (Vals AI) | Not measured | 36.9% | Only GPT-6.2 Sol measured |
| GPQA Diamond (Artificial Analysis run) | 89.5% | Not measured | Only GLM 5.2 measured |
| Humanity's Last Exam (Artificial Analysis run) | 41.1% | Not measured | Only GLM 5.2 measured |
| HLE-Diamond (Artificial Analysis run) | 17.7% | 64.0% | GPT-6.2 Sol |
| CritPt v2, hard subset (Artificial Analysis run) | Not measured | 64.0% | Only GPT-6.2 Sol measured |
| GPQA Diamond (Epoch AI run) | 91.9% | Not measured | Only GLM 5.2 measured |
| ARC-AGI | 77.0% | 97.5% | GPT-6.2 Sol |
| ARC-AGI-2 | 22.8% | 85.8% | GPT-6.2 Sol |
| SimpleBench | 58.8% | 69.9% | GPT-6.2 Sol |
| Chess puzzles | 22.0% | Not measured | Only GLM 5.2 measured |
| Mystery game puzzles | 19.0% | Not measured | Only GLM 5.2 measured |
| EBR-bench | 9.5% | Not measured | Only GLM 5.2 measured |
| LMArena Text: hard prompts | 1481 | 1419 | GLM 5.2 |
| LMArena Text: expert | 1489 | Not measured | Only GLM 5.2 measured |
| LMArena Text: hard prompts in English | 1481 | 1413 | GLM 5.2 |
| Kagi LLM Benchmark | 62.6% | 67.0% | GPT-6.2 Sol |
| Humanity's Last Exam (Scale run) | Not measured | 56.5% | Only GPT-6.2 Sol measured |
| Neither model has a result on the other 7 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| OTIS Mock AIME | 86.4% | Not measured | Only GLM 5.2 measured |
| FrontierMath, tiers 1 to 3 | 59.2% | Not measured | Only GLM 5.2 measured |
| FrontierMath, tier 4 | 29.3% | Not measured | Only GLM 5.2 measured |
| ProofBench | 35.0% | Not measured | Only GLM 5.2 measured |
| LMArena Text: math | 1484 | Not measured | Only GLM 5.2 measured |
| LMArena Text: mathematical work | 1490 | Not measured | Only GLM 5.2 measured |
| MathArena | 20.4% | 91.2% | GPT-6.2 Sol |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index21
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| MMLU Pro | 86.7% | Not measured | Only GLM 5.2 measured |
| LegalBench (Vals AI) | 84.1% | Not measured | Only GLM 5.2 measured |
| CorpFin (Vals AI) | 66.1% | Not measured | Only GLM 5.2 measured |
| TaxEval (Vals AI) | 73.3% | Not measured | Only GLM 5.2 measured |
| MedCode (Vals AI) | 40.8% | Not measured | Only GLM 5.2 measured |
| AA-Omniscience | 0.6 | 40.4 | GPT-6.2 Sol |
| SimpleQA Verified | 34.2% | Not measured | Only GLM 5.2 measured |
| LMArena Text: science | 1489 | 1432 | GLM 5.2 |
| LMArena Text: medicine and healthcare | 1476 | Not measured | Only GLM 5.2 measured |
| LMArena Text: legal and government | 1490 | Not measured | Only GLM 5.2 measured |
| LMArena Text: business and finance | 1450 | 1383 | GLM 5.2 |
| Neither model has a result on the other 10 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| Design Arena: websites | 1272 | Not measured | Only GLM 5.2 measured |
| Design Arena: UI components | 1279 | Not measured | Only GLM 5.2 measured |
| Design Arena: game dev | 1239 | 1437 | GPT-6.2 Sol |
| Design Arena: data viz | 1279 | Not measured | Only GLM 5.2 measured |
| Design Arena: 3D | 1252 | 1482 | GPT-6.2 Sol |
| Design Arena: all code categories | 1268 | 1405 | GPT-6.2 Sol |
| Design Arena: SVG | 1189 | Not measured | Only GLM 5.2 measured |
| Design Arena: ASCII art | 1175 | Not measured | Only GLM 5.2 measured |
| LiveBench design | Not measured | 80.0% | Only GPT-6.2 Sol measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index110
| Benchmark | GLM 5.2 | GPT-6.2 Sol | Leader |
|---|---|---|---|
| Vals Index | Not measured | 66.1% | Only GPT-6.2 Sol measured |
| LiveBench data analysis | Not measured | 81.1% | Only GPT-6.2 Sol measured |
| MedScribe (Vals AI) | 83.5% | Not measured | Only GLM 5.2 measured |
| IFBench | 73.3% | Not measured | Only GLM 5.2 measured |
| Long context reasoning (AA-LCR) | 78.3% | Not measured | Only GLM 5.2 measured |
| GDPval-AA | 1380 | Not measured | Only GLM 5.2 measured |
| DTBench | 93.6% | Not measured | Only GLM 5.2 measured |
| LMCA | 45.8% | Not measured | Only GLM 5.2 measured |
| Surface Evolver Bench | 55.6% | Not measured | Only GLM 5.2 measured |
| LMArena Text | 1469 | 1414 | GLM 5.2 |
| LMArena Text: instruction following | 1467 | 1434 | GLM 5.2 |
| LMArena Text: creative writing | 1460 | 1403 | GLM 5.2 |
| LMArena Text: English | 1475 | 1419 | GLM 5.2 |
| LMArena Text: non-English | 1458 | 1405 | GLM 5.2 |
| LMArena Text: Chinese | 1523 | 1434 | GLM 5.2 |
| LMArena Text: German | 1467 | Not measured | Only GLM 5.2 measured |
| LMArena Text: Japanese | 1453 | Not measured | Only GLM 5.2 measured |
| LMArena Text: Korean | 1429 | Not measured | Only GLM 5.2 measured |
| LMArena Text: French | 1476 | Not measured | Only GLM 5.2 measured |
| LMArena Text: Spanish | 1482 | Not measured | Only GLM 5.2 measured |
| LMArena Text: Russian | 1466 | 1401 | GLM 5.2 |
| LMArena Text: Polish | 1459 | Not measured | Only GLM 5.2 measured |
| LMArena Text: multi-turn | 1470 | 1420 | GLM 5.2 |
| LMArena Text: longer queries | 1479 | 1427 | GLM 5.2 |
| LMArena Text: writing and language | 1461 | 1420 | GLM 5.2 |
| LMArena Text: entertainment and media | 1448 | 1408 | GLM 5.2 |
| MultiChallenge | Not measured | 72.0% | Only GPT-6.2 Sol measured |
| EQ-Bench 4 | 1208 | Not measured | Only GLM 5.2 measured |
| Neither model has a result on the other 21 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GLM 5.2 | GPT-6.2 Sol | 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 5.2 costs less
Repository review
50K input + 3K output tokens
GLM 5.2 costs less
Agent loop
200K input + 10K output tokens
GLM 5.2 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 5.2 | OpenAI GPT-6.2 Sol |
|---|---|---|
| Maker | Z.ai | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | June 16, 2026 | November 17, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $1.40 | $2.00 |
| Output $/Mtok | $4.40 | $10.00 |
| Coding index | 61.1 | 83.1 |
| Independent measurements behind it | 4 | 4 |
| Cost per solved task | $0.86 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GLM 5.2 better than GPT-6.2 Sol for coding?
Pick GPT-6.2 Sol for coding: it scores 83.1 on the Vibe Coding Index against 61.1 for GLM 5.2. Pick GLM 5.2 for output tokens 2.3x cheaper ($4.40 vs $10.00 per million) and work measured by LMArena Text: coding or LMArena Text: software and IT, where it leads.
Which is cheaper, GLM 5.2 or GPT-6.2 Sol?
GLM 5.2 costs $1.40 per million input tokens and $4.40 per million output tokens; GPT-6.2 Sol costs $2.00 and $10.00. Cost per solved task is known for GLM 5.2 only.
Which has the bigger context window, GLM 5.2 or GPT-6.2 Sol?
GPT-6.2 Sol, with 1.05M tokens against 1.05M.
How many benchmarks is the GLM 5.2 vs GPT-6.2 Sol comparison based on?
35 benchmarks have a result for both models as of November 22, 2026. GLM 5.2's coding index rests on 4 independent benchmarks and GPT-6.2 Sol's on 4. Where only one model has a result, the page says so instead of naming a winner.