Model comparison
DeepSeek-V2.5 (Sep 2024) vs GPT-5.6 Terra
GPT-5.6 Terra is the stronger model overall, scoring 59.1 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 GPT-5.6 Terra in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Terra leads 81.5 to 33.3.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.0 | 59.1 |
| Released | 2024-09-06 | 2026-07-09 |
| Weights | Open | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 128K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $12 |
| Results tracked | 22 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 31.3 (#274), GPT-5.6 Terra: 57.1 (#16)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Coding | 1310 | 1484 |
| DeepSWE | — | 69.6% |
| FrontierCode | — | 41.3% |
| Aider Polyglot | 17.8% | — |
| CursorBench | — | 41.3% |
| LMArena WebDev | — | 1520 |
| SciCode | — | 55% |
| WeirdML | — | 78.3% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 1,951 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5.6 Terra: 40.2 (#23)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| APEX-Agents | — | 58.2% |
| BALROG | — | 53.2% |
| GDP.pdf | — | 24.7% |
| Vending-Bench 2 | — | 7,343 |
Reasoning GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 24.5 (#145), GPT-5.6 Terra: 60.6 (#19)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1467 |
| ARC-AGI-2 | — | 83.9% |
| SimpleBench | — | 48.9% |
| Kagi LLM Benchmark | — | 51.3% |
| NYT Connections (extended) | — | 78.4% |
| ARC-AGI-1 | — | 96.5% |
| CritPt | — | 30% |
| Chess Puzzles | — | 54% |
| Mystery Game Puzzles | — | 35% |
| DTBench | — | 93.3% |
| LMCA | — | 55% |
| Surface Evolver Bench | — | 83.8% |
| Epoch Capabilities Index | — | 159.31 |
Math GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 33.3 (#177), GPT-5.6 Terra: 81.5 (#10)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Math | 1289 | 1463 |
| FrontierMath (Tiers 1-3) | — | 86% |
| FrontierMath Tier 4 | — | 70.7% |
| OTIS Mock AIME 2024-2025 | — | 99.7% |
| ProofBench | — | 74% |
Knowledge GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 34.2 (#190), GPT-5.6 Terra: 61.3 (#27)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Expert | 1266 | 1492 |
| GPQA Diamond | — | 93.3% |
| SimpleQA Verified | — | 43.2% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, GPT-5.6 Terra: 47.5 (#10)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Vision | — | 1271 |
| Blueprint-Bench 2 | — | 30.8% |
| Furniture Assembly | — | 54.2% |
| LMArena Document | — | 1472 |
Multilingual GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 42.7 (#188), GPT-5.6 Terra: 54.5 (#42)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Non-English | 1273 | 1439 |
| LMArena Chinese | 1317 | 1514 |
| LMArena French | 1288 | 1476 |
| LMArena German | 1258 | 1462 |
| LMArena Japanese | 1228 | 1446 |
| LMArena Korean | 1209 | 1420 |
| LMArena Russian | 1289 | 1450 |
| LMArena Spanish | 1248 | 1455 |
Instruction Following GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 67.7 (#189), GPT-5.6 Terra: 76.6 (#38)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Instruction Following | 1280 | 1453 |
Long Context GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#169), GPT-5.6 Terra: 44.4 (#64)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Longer Query | 1301 | 1452 |
Writing & Preference GPT-5.6 Terra leads
DeepSeek-V2.5 (Sep 2024): 50.0 (#182), GPT-5.6 Terra: 70.4 (#21)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | GPT-5.6 Terra |
|---|---|---|
| LMArena Text | 1294 | 1446 |
| LMArena Creative Writing | 1285 | 1411 |
| LMArena Multi-Turn | 1297 | 1451 |
| EQ-Bench Creative Writing | — | 1855 |
| EQ-Bench 4 | — | 1234 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than GPT-5.6 Terra?
GPT-5.6 Terra is the stronger model overall, scoring 59.1 to 37.0 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or GPT-5.6 Terra better for coding?
GPT-5.6 Terra scores higher on coding benchmarks: 57.1 versus 31.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and GPT-5.6 Terra share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and GPT-5.6 Terra has 52.