Model comparison
DeepSeek-V3 vs GLM-5V-Turbo
GLM-5V-Turbo is the stronger model overall, scoring 43.8 to 39.5 on the Noometry Index. DeepSeek-V3 costs 4.7× less per token, which makes it the better buy when GLM-5V-Turbo's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
DeepSeekSummary
- They share 16 benchmarks with published results for both. DeepSeek-V3 scores higher in 1 category and GLM-5V-Turbo in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where GLM-5V-Turbo leads 44.0 to 34.0.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $1.20 / $4 for GLM-5V-Turbo.
- GLM-5V-Turbo accepts more context: 200K tokens versus 164K.
- DeepSeek-V3 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3 | GLM-5V-Turbo | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 39.5 | 43.8 |
| Released | 2024-12-26 | 2026-04-01 |
| Weights | Open | Proprietary |
| Context window | 164K | 200K |
| Max output | 164K | 131K |
| Input $ / M tokens | $0.24 | $1.20 |
| Output $ / M tokens | $0.90 | $4 |
| Results tracked | 60 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3: 42.3 (#106), GLM-5V-Turbo: 42.1 (#111)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Coding | 1368 | 1466 |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1401 |
| SciCode | 35.8% | — |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, GLM-5V-Turbo: —
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning GLM-5V-Turbo leads
DeepSeek-V3: 20.5 (#236), GLM-5V-Turbo: 29.7 (#89)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Hard Prompts | 1365 | 1443 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | 65.8% | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| Epoch Capabilities Index | 135.94 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |
Math GLM-5V-Turbo leads
DeepSeek-V3: 32.1 (#219), GLM-5V-Turbo: 39.4 (#106)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Math | 1373 | 1441 |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge GLM-5V-Turbo leads
DeepSeek-V3: 37.5 (#155), GLM-5V-Turbo: 40.6 (#117)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Expert | 1351 | 1452 |
| GPQA Diamond | 67.6% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |
Multimodal Not comparable
DeepSeek-V3: —, GLM-5V-Turbo: 40.9 (#42)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Vision | — | 1264 |
| LMArena Document | — | 1416 |
Multilingual GLM-5V-Turbo leads
DeepSeek-V3: 48.5 (#143), GLM-5V-Turbo: 53.0 (#73)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Non-English | 1358 | 1420 |
| LMArena Chinese | 1391 | 1488 |
| LMArena French | 1385 | 1444 |
| LMArena German | 1374 | 1423 |
| LMArena Korean | 1319 | 1396 |
| LMArena Russian | 1373 | 1431 |
| LMArena Spanish | 1358 | 1450 |
| LMArena Japanese | 1333 | — |
Instruction Following GLM-5V-Turbo leads
DeepSeek-V3: 72.8 (#130), GLM-5V-Turbo: 75.0 (#80)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Instruction Following | 1345 | 1423 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
Long Context GLM-5V-Turbo leads
DeepSeek-V3: 34.0 (#253), GLM-5V-Turbo: 44.0 (#80)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Longer Query | 1352 | 1438 |
| Fiction.LiveBench | 50% | — |
Writing & Preference GLM-5V-Turbo leads
DeepSeek-V3: 57.4 (#130), GLM-5V-Turbo: 62.5 (#73)
| Benchmark | DeepSeek-V3 | GLM-5V-Turbo |
|---|---|---|
| LMArena Text | 1375 | 1437 |
| LMArena Creative Writing | 1364 | 1416 |
| LMArena Multi-Turn | 1389 | 1432 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than GLM-5V-Turbo?
GLM-5V-Turbo is the stronger model overall, scoring 43.8 to 39.5 on the Noometry Index. DeepSeek-V3 costs 4.7× less per token, which makes it the better buy when GLM-5V-Turbo's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3 or GLM-5V-Turbo?
DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; GLM-5V-Turbo lists at $1.20 and $4.
Is DeepSeek-V3 or GLM-5V-Turbo better for coding?
They score almost the same on coding (42.3 vs 42.1); test both on your own repository before choosing.
Which has the bigger context window?
GLM-5V-Turbo does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-V3 and GLM-5V-Turbo share?
16 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and GLM-5V-Turbo has 19.