Model comparison
DeepSeek-V2.5 (Sep 2024) vs GLM-5.2
GLM-5.2 is the stronger model overall, scoring 50.9 to 37.0 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 0 categories and GLM-5.2 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.2 leads 57.1 to 34.2.
Side by side
| DeepSeek-V2.5 (Sep 2024) | GLM-5.2 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 37.0 | 50.9 |
| Released | 2024-09-06 | 2026-06-13 |
| Weights | Open | Open |
| Context window | — | 1M |
| Max output | — | 131K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $4.40 |
| Results tracked | 22 | 50 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 31.3 (#274), GLM-5.2: 50.9 (#37)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Coding | 1310 | 1484 |
| SWE-bench Verified | — | 78.7% |
| DeepSWE | — | 43.8% |
| FrontierCode | — | 24.5% |
| Aider Polyglot | 17.8% | — |
| LMArena WebDev | — | 1604 |
| SciCode | — | 50.5% |
| WeirdML | — | 70.1% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 1,047 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, GLM-5.2: 31.6 (#63)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 31.7% |
| GBAEval | — | 0% |
| Vending-Bench 2 | — | 8,314 |
Reasoning GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 24.5 (#145), GLM-5.2: 42.1 (#46)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1480 |
| ARC-AGI-2 | — | 22.8% |
| SimpleBench | — | 58.8% |
| Kagi LLM Benchmark | — | 62.6% |
| NYT Connections (extended) | — | 74.3% |
| ARC-AGI-1 | — | 77% |
| CritPt | — | 20.9% |
| Chess Puzzles | — | 21% |
| EBR-Bench | — | 9.5% |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 93.6% |
| LMCA | — | 45.8% |
| Surface Evolver Bench | — | 55.6% |
| Epoch Capabilities Index | — | 151.76 |
Math GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 33.3 (#177), GLM-5.2: 55.4 (#35)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Math | 1289 | 1482 |
| FrontierMath (Tiers 1-3) | — | 59.2% |
| FrontierMath Tier 4 | — | 29.3% |
| MathArena Final-Answer Competitions | — | 67.6% |
| OTIS Mock AIME 2024-2025 | — | 86.4% |
| ProofBench | — | 35% |
Knowledge GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 34.2 (#190), GLM-5.2: 57.1 (#36)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Expert | 1266 | 1484 |
| GPQA Diamond | — | 91.9% |
| SimpleQA Verified | — | 34.2% |
Multilingual GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 42.7 (#188), GLM-5.2: 56.0 (#25)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Non-English | 1273 | 1459 |
| LMArena Chinese | 1317 | 1518 |
| LMArena French | 1288 | 1476 |
| LMArena German | 1258 | 1470 |
| LMArena Japanese | 1228 | 1451 |
| LMArena Korean | 1209 | 1445 |
| LMArena Russian | 1289 | 1467 |
| LMArena Spanish | 1248 | 1477 |
Instruction Following GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 67.7 (#189), GLM-5.2: 77.1 (#32)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Instruction Following | 1280 | 1465 |
Long Context GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#169), GLM-5.2: 45.3 (#43)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Longer Query | 1301 | 1479 |
Writing & Preference GLM-5.2 leads
DeepSeek-V2.5 (Sep 2024): 50.0 (#182), GLM-5.2: 70.7 (#20)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GLM-5.2 |
|---|---|---|
| LMArena Text | 1294 | 1470 |
| LMArena Creative Writing | 1285 | 1465 |
| LMArena Multi-Turn | 1297 | 1468 |
| EQ-Bench Creative Writing | — | 1757 |
| EQ-Bench 4 | — | 1222 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than GLM-5.2?
GLM-5.2 is the stronger model overall, scoring 50.9 to 37.0 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or GLM-5.2 better for coding?
GLM-5.2 scores higher on coding benchmarks: 50.9 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and GLM-5.2 share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and GLM-5.2 has 50.