Model comparison
DeepSeek-V3.2-Exp vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 51.1 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 7.4× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
DeepSeekSummary
- They share 36 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and GLM-5.2 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.2 leads 42.3 to 22.1.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 74.3% for GLM-5.2.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $1.40 / $4.40 for GLM-5.2.
- GLM-5.2 accepts more context: 1M tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | GLM-5.2 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 44.3 | 51.1 |
| Released | 2025-09-29 | 2026-06-13 |
| Weights | Open | Open |
| Context window | 164K | 1M |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.26 | $1.40 |
| Output $ / M tokens | $0.38 | $4.40 |
| Results tracked | 49 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
DeepSeek-V3.2-Exp: 46.5 (#65), GLM-5.2: 51.3 (#41)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| LMArena WebDev | 1362 | 1603 |
| SciCode | 38.9% | 50.5% |
| WeirdML | 39.5% | 70.1% |
| LMArena Coding | 1454 | 1485 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| ALE-Bench | — | 1,047 |
Agentic & Tool Use Too close to call
DeepSeek-V3.2-Exp: 32.7 (#59), GLM-5.2: 32.4 (#63)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| APEX-Agents | 21.3% | 45.2% |
| Vending-Bench 2 | 1,034 | 8,314 |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
Reasoning GLM-5.2 leads
DeepSeek-V3.2-Exp: 22.1 (#208), GLM-5.2: 42.3 (#52)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| ARC-AGI-2 | 4% | 22.8% |
| Kagi LLM Benchmark | 52.2% | 62.6% |
| NYT Connections (extended) | 36.7% | 74.3% |
| ARC-AGI-1 | 57% | 77% |
| CritPt | 2.9% | 20.9% |
| Chess Puzzles | 14% | 21% |
| LMArena Hard Prompts | 1434 | 1480 |
| DTBench | 87.7% | 93.6% |
| LMCA | 29.1% | 45.8% |
| Epoch Capabilities Index | 146.27 | 151.78 |
| SimpleBench | — | 58.8% |
| Thematic Generalization | 65% | — |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| Surface Evolver Bench | — | 55.6% |
Math GLM-5.2 leads
DeepSeek-V3.2-Exp: 41.7 (#87), GLM-5.2: 55.7 (#43)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| MathArena Final-Answer Competitions | 57.7% | 67.6% |
| OTIS Mock AIME 2024-2025 | 87.8% | 86.4% |
| ProofBench | 8% | 35% |
| LMArena Math | 1435 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GLM-5.2 leads
DeepSeek-V3.2-Exp: 51.7 (#66), GLM-5.2: 57.1 (#40)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| GPQA Diamond | 83.4% | 91.9% |
| LMArena Expert | 1436 | 1486 |
| SimpleQA Verified | — | 34.2% |
| Vectara Hallucination Rate | 5.3% | — |
Multilingual GLM-5.2 leads
DeepSeek-V3.2-Exp: 52.2 (#90), GLM-5.2: 55.8 (#26)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1409 | 1459 |
| LMArena Chinese | 1461 | 1519 |
| LMArena French | 1433 | 1479 |
| LMArena German | 1440 | 1468 |
| LMArena Japanese | 1374 | 1451 |
| LMArena Korean | 1371 | 1445 |
| LMArena Russian | 1424 | 1466 |
| LMArena Spanish | 1440 | 1477 |
Instruction Following GLM-5.2 leads
DeepSeek-V3.2-Exp: 74.5 (#93), GLM-5.2: 76.9 (#34)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1413 | 1465 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), GLM-5.2: 45.3 (#43)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1428 | 1479 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference GLM-5.2 leads
DeepSeek-V3.2-Exp: 62.4 (#77), GLM-5.2: 70.4 (#21)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.2 |
|---|---|---|
| LMArena Text | 1425 | 1470 |
| LMArena Creative Writing | 1403 | 1462 |
| EQ-Bench Creative Writing | 1515 | 1757 |
| LMArena Multi-Turn | 1427 | 1469 |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 51.1 to 44.3 on the Noometry Index. DeepSeek-V3.2-Exp costs 7.4× less per token, which makes it the better buy when GLM-5.2's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.2-Exp or GLM-5.2?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; GLM-5.2 lists at $1.40 and $4.40.
Is DeepSeek-V3.2-Exp or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 51.3 versus 46.5 in the Noometry coding category.
Which has the bigger context window?
GLM-5.2 does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and GLM-5.2 share?
36 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and GLM-5.2 has 51.