Model comparison
DeepSeek-V3.1 vs GLM-5
GLM-5 is the stronger model overall, scoring 46.1 to 42.8 on the Noometry Index. DeepSeek-V3.1 costs 3.6× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
DeepSeekSummary
- They share 24 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 1 category and GLM-5 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where GLM-5 leads 49.0 to 40.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for DeepSeek-V3.1 and 75% for GLM-5.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $1 / $3.20 for GLM-5.
- GLM-5 accepts more context: 205K tokens versus 164K.
Side by side
| DeepSeek-V3.1 | GLM-5 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 42.8 | 46.1 |
| Released | 2025-08-21 | 2026-02-11 |
| Weights | Open | Open |
| Context window | 164K | 205K |
| Max output | 8K | 131K |
| Input $ / M tokens | $0.25 | $1 |
| Output $ / M tokens | $0.95 | $3.20 |
| Results tracked | 27 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5 leads
DeepSeek-V3.1: 40.3 (#144), GLM-5: 49.0 (#52)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| WeirdML | 38.4% | 48.2% |
| LMArena Coding | 1417 | 1461 |
| SWE-bench Verified | — | 72.1% |
| SWE-bench Verified (bash only) | — | 72.8% |
| LMArena WebDev | — | 1434 |
| SWE-bench Multilingual | — | 69.7% |
| ALE-Bench | — | 765.62 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, GLM-5: 31.1 (#71)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| Terminal-Bench | — | 52.4% |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 73.7% |
| τ²-bench Telecom | — | 86.8% |
| Vending-Bench 2 | — | 4,432 |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), GLM-5: 27.6 (#116)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| SimpleBench | 40% | 53.2% |
| Kagi LLM Benchmark | 53.2% | 75% |
| LMArena Hard Prompts | 1417 | 1452 |
| Epoch Capabilities Index | 139.92 | 145.83 |
| ForecastBench | 58 | 61 |
| ARC-AGI-2 | — | 4.9% |
| NYT Connections (extended) | — | 74.8% |
| ARC-AGI-1 | — | 44.7% |
| Chess Puzzles | — | 10% |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
Math GLM-5 leads
DeepSeek-V3.1: 38.9 (#122), GLM-5: 46.4 (#71)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| LMArena Math | 1420 | 1440 |
| MathArena Final-Answer Competitions | — | 65.7% |
| OTIS Mock AIME 2024-2025 | — | 80% |
| FrontierMath (Feb 2025 set) | — | 16.4% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge GLM-5 leads
DeepSeek-V3.1: 43.7 (#90), GLM-5: 52.3 (#64)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 10.1% |
| LMArena Expert | 1405 | 1454 |
| GPQA Diamond | — | 87.8% |
Multilingual GLM-5 leads
DeepSeek-V3.1: 51.6 (#106), GLM-5: 53.7 (#58)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| LMArena Non-English | 1400 | 1430 |
| LMArena Chinese | 1469 | 1511 |
| LMArena French | 1447 | 1455 |
| LMArena German | 1411 | 1445 |
| LMArena Japanese | 1378 | 1416 |
| LMArena Korean | 1337 | 1423 |
| LMArena Russian | 1405 | 1436 |
| LMArena Spanish | 1431 | 1454 |
Instruction Following GLM-5 leads
DeepSeek-V3.1: 73.9 (#110), GLM-5: 75.2 (#67)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1428 |
Long Context GLM-5 leads
DeepSeek-V3.1: 36.3 (#232), GLM-5: 44.7 (#60)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| LMArena Longer Query | 1422 | 1446 |
| Fiction.LiveBench | 52.8% | — |
| CL-bench | — | 18.7% |
Writing & Preference GLM-5 leads
DeepSeek-V3.1: 60.3 (#98), GLM-5: 66.0 (#38)
| Benchmark | DeepSeek-V3.1 | GLM-5 |
|---|---|---|
| LMArena Text | 1420 | 1446 |
| LMArena Creative Writing | 1401 | 1439 |
| EQ-Bench Creative Writing | 1436 | 1601 |
| LMArena Multi-Turn | 1408 | 1456 |
Frequently asked questions
Is DeepSeek-V3.1 better than GLM-5?
GLM-5 is the stronger model overall, scoring 46.1 to 42.8 on the Noometry Index. DeepSeek-V3.1 costs 3.6× less per token, which makes it the better buy when GLM-5's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1 or GLM-5?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; GLM-5 lists at $1 and $3.20.
Is DeepSeek-V3.1 or GLM-5 better for coding?
GLM-5 scores higher on coding benchmarks: 49.0 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
GLM-5 does, with 205K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and GLM-5 share?
24 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and GLM-5 has 45.