Model comparison
DeepSeek-V3.2-Exp vs GLM-5.1
GLM-5.1 is the stronger model overall, scoring 47.8 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-5.1's lead doesn't matter for your workload.
Last verified . 34 shared benchmarks.
DeepSeekSummary
- They share 34 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and GLM-5.1 in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-5.1 leads 39.1 to 22.1.
- The biggest single-benchmark swing is NYT Connections (extended): 36.7% for DeepSeek-V3.2-Exp and 77.7% for GLM-5.1.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $1.40 / $4.40 for GLM-5.1.
- GLM-5.1 accepts more context: 200K tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | GLM-5.1 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 44.3 | 47.8 |
| Released | 2025-09-29 | 2026-04-07 |
| Weights | Open | Open |
| Context window | 164K | 200K |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.26 | $1.40 |
| Output $ / M tokens | $0.38 | $4.40 |
| Results tracked | 49 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
DeepSeek-V3.2-Exp: 46.5 (#65), GLM-5.1: 48.7 (#55)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| LMArena WebDev | 1362 | 1508 |
| SciCode | 38.9% | 43.8% |
| WeirdML | 39.5% | 57.1% |
| LMArena Coding | 1454 | 1485 |
| SWE-bench Verified | — | 74.2% |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| ALE-Bench | — | 887.1 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 32.7 (#59), GLM-5.1: 24.9 (#113)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| APEX-Agents | 21.3% | 40.9% |
| Vending-Bench 2 | 1,034 | 5,634 |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| ExploitBench | — | 18.1% |
| GBAEval | — | 0% |
Reasoning GLM-5.1 leads
DeepSeek-V3.2-Exp: 22.1 (#208), GLM-5.1: 39.1 (#60)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| NYT Connections (extended) | 36.7% | 77.7% |
| CritPt | 2.9% | 4.6% |
| Chess Puzzles | 14% | 19% |
| Thematic Generalization | 65% | 69.8% |
| LMArena Hard Prompts | 1434 | 1472 |
| Epoch Capabilities Index | 146.27 | 149.84 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 55.1% |
| Kagi LLM Benchmark | 52.2% | — |
| ARC-AGI-1 | 57% | — |
| DTBench | 87.7% | — |
| LMCA | 29.1% | — |
Math GLM-5.1 leads
DeepSeek-V3.2-Exp: 41.7 (#87), GLM-5.1: 49.7 (#60)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| MathArena Final-Answer Competitions | 57.7% | 67.1% |
| OTIS Mock AIME 2024-2025 | 87.8% | 93.3% |
| ProofBench | 8% | 22.2% |
| LMArena Math | 1435 | 1473 |
| FrontierMath (Feb 2025 set) | 22.1% | 33.4% |
| FrontierMath Tier 4 (v1) | 2.1% | 12.5% |
| FrontierMath (Tiers 1-3) | — | 36.8% |
Knowledge GLM-5.1 leads
DeepSeek-V3.2-Exp: 51.7 (#66), GLM-5.1: 54.9 (#50)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| GPQA Diamond | 83.4% | 89.9% |
| LMArena Expert | 1436 | 1476 |
| SimpleQA Verified | — | 34% |
| Vectara Hallucination Rate | 5.3% | — |
Multilingual GLM-5.1 leads
DeepSeek-V3.2-Exp: 52.2 (#90), GLM-5.1: 55.0 (#36)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| LMArena Non-English | 1409 | 1447 |
| LMArena Chinese | 1461 | 1515 |
| LMArena French | 1433 | 1474 |
| LMArena German | 1440 | 1465 |
| LMArena Japanese | 1374 | 1434 |
| LMArena Korean | 1371 | 1418 |
| LMArena Russian | 1424 | 1454 |
| LMArena Spanish | 1440 | 1469 |
Instruction Following GLM-5.1 leads
DeepSeek-V3.2-Exp: 74.5 (#93), GLM-5.1: 76.3 (#42)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| LMArena Instruction Following | 1413 | 1451 |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#16), GLM-5.1: 44.9 (#53)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| LMArena Longer Query | 1428 | 1466 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference GLM-5.1 leads
DeepSeek-V3.2-Exp: 62.4 (#77), GLM-5.1: 66.9 (#31)
| Benchmark | DeepSeek-V3.2-Exp | GLM-5.1 |
|---|---|---|
| LMArena Text | 1425 | 1461 |
| LMArena Creative Writing | 1403 | 1453 |
| EQ-Bench Creative Writing | 1515 | 1592 |
| LMArena Multi-Turn | 1427 | 1472 |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than GLM-5.1?
GLM-5.1 is the stronger model overall, scoring 47.8 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-5.1's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.2-Exp or GLM-5.1?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; GLM-5.1 lists at $1.40 and $4.40.
Is DeepSeek-V3.2-Exp or GLM-5.1 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.7 versus 46.5 in the Noometry coding category.
Which has the bigger context window?
GLM-5.1 does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and GLM-5.1 share?
34 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and GLM-5.1 has 41.