Model comparison
DeepSeek-V3.1-Terminus vs GLM-5.1
GLM-5.1 is the stronger model overall, scoring 47.6 to 42.6 on the Noometry Index. DeepSeek-V3.1-Terminus costs 4.5× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
DeepSeekSummary
- They share 13 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and GLM-5.1 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.1 leads 49.0 to 35.7.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 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-Terminus | GLM-5.1 | |
|---|---|---|
| Provider | DeepSeek | Z.ai (Zhipu) |
| Noometry Index | 42.6 | 47.6 |
| Released | 2025-09-22 | 2026-04-07 |
| Weights | Open | Open |
| Context window | 164K | 200K |
| Max output | 66K | 131K |
| Input $ / M tokens | $0.30 | $1.40 |
| Output $ / M tokens | $1 | $4.40 |
| Results tracked | 16 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.1 leads
DeepSeek-V3.1-Terminus: 41.5 (#114), GLM-5.1: 48.5 (#52)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| SciCode | 40.6% | 43.8% |
| LMArena Coding | 1426 | 1486 |
| ALE-Bench | 745.17 | 887.1 |
| SWE-bench Verified | — | 74.2% |
| LMArena WebDev | — | 1509 |
| WeirdML | — | 57.1% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, GLM-5.1: 24.8 (#106)
| Benchmark | DeepSeek-V3.1-Terminus | 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-Terminus: 25.7 (#126), GLM-5.1: 38.2 (#57)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| CritPt | 1.7% | 4.6% |
| LMArena Hard Prompts | 1427 | 1472 |
| SimpleBench | — | 55.1% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 77.7% |
| Chess Puzzles | — | 19% |
| Thematic Generalization | — | 69.8% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 149.86 |
Math GLM-5.1 leads
DeepSeek-V3.1-Terminus: 35.7 (#138), GLM-5.1: 49.0 (#48)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| LMArena Math | 1402 | 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 Not comparable
DeepSeek-V3.1-Terminus: —, GLM-5.1: 55.0 (#45)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| GPQA Diamond | — | 89.9% |
| SimpleQA Verified | — | 34% |
| LMArena Expert | — | 1477 |
Multilingual GLM-5.1 leads
DeepSeek-V3.1-Terminus: 52.3 (#87), GLM-5.1: 55.2 (#34)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| LMArena Non-English | 1407 | 1448 |
| LMArena Russian | 1434 | 1453 |
| LMArena Chinese | — | 1516 |
| LMArena French | — | 1478 |
| LMArena German | — | 1466 |
| LMArena Japanese | — | 1436 |
| LMArena Korean | — | 1423 |
| LMArena Spanish | — | 1467 |
Instruction Following GLM-5.1 leads
DeepSeek-V3.1-Terminus: 74.4 (#96), GLM-5.1: 76.5 (#41)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1451 |
Long Context GLM-5.1 leads
DeepSeek-V3.1-Terminus: 43.5 (#88), GLM-5.1: 44.9 (#51)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| LMArena Longer Query | 1426 | 1467 |
Writing & Preference GLM-5.1 leads
DeepSeek-V3.1-Terminus: 61.3 (#87), GLM-5.1: 67.2 (#29)
| Benchmark | DeepSeek-V3.1-Terminus | GLM-5.1 |
|---|---|---|
| LMArena Text | 1420 | 1462 |
| LMArena Creative Writing | 1404 | 1454 |
| LMArena Multi-Turn | 1411 | 1473 |
| EQ-Bench Creative Writing | — | 1592 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than GLM-5.1?
GLM-5.1 is the stronger model overall, scoring 47.6 to 42.6 on the Noometry Index. DeepSeek-V3.1-Terminus costs 4.5× 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-Terminus or GLM-5.1?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; GLM-5.1 lists at $1.40 and $4.40.
Is DeepSeek-V3.1-Terminus or GLM-5.1 better for coding?
GLM-5.1 scores higher on coding benchmarks: 48.5 versus 41.5 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-V3.1-Terminus and GLM-5.1 share?
13 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and GLM-5.1 has 41.