Model comparison
DeepSeek-V3.1 vs GLM-5.1
GLM-5.1 is the stronger model overall, scoring 47.6 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 5.1× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
DeepSeekSummary
- They share 21 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 0 categories and GLM-5.1 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.1 leads 49.0 to 36.1.
- The biggest single-benchmark swing is WeirdML: 38.4% for DeepSeek-V3.1 and 57.1% for GLM-5.1.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 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-V3.1 | GLM-5.1 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 42.2 | 47.6 |
| Released | 2025-08-21 | 2026-04-07 |
| Weights | Open | Open |
| Context window | 164K | 200K |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.25 | $1.40 |
| Output $ / M tokens | $0.95 | $4.40 |
| Results tracked | 27 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
DeepSeek-V3.1: 39.8 (#140), GLM-5.1: 48.5 (#52)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| WeirdML | 38.4% | 57.1% |
| LMArena Coding | 1417 | 1486 |
| SWE-bench Verified | — | 74.2% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 43.8% |
| ALE-Bench | — | 887.1 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, GLM-5.1: 24.8 (#106)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| APEX-Agents | — | 40.9% |
| ExploitBench | — | 18.1% |
| GBAEval | — | 0% |
| Vending-Bench 2 | — | 5,634 |
Reasoning GLM-5.1 leads
DeepSeek-V3.1: 27.3 (#109), GLM-5.1: 38.2 (#57)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| SimpleBench | 40% | 55.1% |
| LMArena Hard Prompts | 1417 | 1472 |
| Epoch Capabilities Index | 139.92 | 149.86 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 77.7% |
| CritPt | — | 4.6% |
| Chess Puzzles | — | 19% |
| Thematic Generalization | — | 69.8% |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| ForecastBench | 58 | — |
Math GLM-5.1 leads
DeepSeek-V3.1: 36.1 (#125), GLM-5.1: 49.0 (#48)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Math | 1420 | 1472 |
| FrontierMath (Tiers 1-3) | — | 36.8% |
| MathArena Final-Answer Competitions | — | 67.1% |
| OTIS Mock AIME 2024-2025 | — | 93.3% |
| ProofBench | — | 22.2% |
| FrontierMath (Feb 2025 set) | — | 33.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge GLM-5.1 leads
DeepSeek-V3.1: 43.0 (#86), GLM-5.1: 55.0 (#45)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Expert | 1403 | 1477 |
| GPQA Diamond | — | 89.9% |
| SimpleQA Verified | — | 34% |
| Vectara Hallucination Rate | 5.5% | — |
Multilingual GLM-5.1 leads
DeepSeek-V3.1: 51.8 (#100), GLM-5.1: 55.2 (#34)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Non-English | 1400 | 1448 |
| LMArena Chinese | 1470 | 1516 |
| LMArena French | 1447 | 1478 |
| LMArena German | 1412 | 1466 |
| LMArena Japanese | 1374 | 1436 |
| LMArena Korean | 1337 | 1423 |
| LMArena Russian | 1405 | 1453 |
| LMArena Spanish | 1429 | 1467 |
Instruction Following GLM-5.1 leads
DeepSeek-V3.1: 74.1 (#105), GLM-5.1: 76.5 (#41)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1451 |
Long Context GLM-5.1 leads
DeepSeek-V3.1: 36.3 (#227), GLM-5.1: 44.9 (#51)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Longer Query | 1422 | 1467 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference GLM-5.1 leads
DeepSeek-V3.1: 60.5 (#93), GLM-5.1: 67.2 (#29)
| Benchmark | DeepSeek-V3.1 | GLM-5.1 |
|---|---|---|
| LMArena Text | 1420 | 1462 |
| LMArena Creative Writing | 1401 | 1454 |
| EQ-Bench Creative Writing | 1436 | 1592 |
| LMArena Multi-Turn | 1408 | 1473 |
Frequently asked questions
Is DeepSeek-V3.1 better than GLM-5.1?
GLM-5.1 is the stronger model overall, scoring 47.6 to 42.2 on the Noometry Index. DeepSeek-V3.1 costs 5.1× 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-V3.1 or GLM-5.1?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; GLM-5.1 lists at $1.40 and $4.40.
Is DeepSeek-V3.1 or GLM-5.1 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.5 versus 39.8 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-V3.1 and GLM-5.1 share?
21 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and GLM-5.1 has 41.