Model comparison
DeepSeek-V2.5 (Sep 2024) vs Nova 2 Lite
Nova 2 Lite is the stronger model overall, scoring 39.7 to 37.6 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 Nova 2 Lite in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Nova 2 Lite leads 40.7 to 31.7.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite | |
|---|---|---|
| Provider | DeepSeek | Amazon |
| Noometry Index | 37.6 | 39.7 |
| Released | 2024-09-06 | 2025-12-01 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 64K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $2.50 |
| Results tracked | 22 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Nova 2 Lite: 40.7 (#134)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1309 | 1385 |
| Aider Polyglot | 17.8% | — |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Nova 2 Lite: 24.1 (#121)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |
Reasoning Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Nova 2 Lite: 27.5 (#118)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1364 |
Math Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Nova 2 Lite: 37.5 (#156)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1288 | 1359 |
Knowledge Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Nova 2 Lite: 43.0 (#94)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Expert | 1266 | 1358 |
| Vectara Hallucination Rate | — | 5.1% |
Multilingual Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Nova 2 Lite: 47.1 (#153)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1273 | 1337 |
| LMArena Chinese | 1318 | 1364 |
| LMArena French | 1289 | 1381 |
| LMArena German | 1258 | 1343 |
| LMArena Japanese | 1228 | 1271 |
| LMArena Korean | 1209 | 1284 |
| LMArena Russian | 1289 | 1343 |
| LMArena Spanish | 1248 | 1373 |
Instruction Following Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Nova 2 Lite: 70.5 (#161)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1280 | 1335 |
Long Context Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Nova 2 Lite: 40.6 (#150)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1301 | 1335 |
Writing & Preference Nova 2 Lite leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Nova 2 Lite: 53.9 (#154)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1294 | 1362 |
| LMArena Creative Writing | 1285 | 1291 |
| LMArena Multi-Turn | 1297 | 1338 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Nova 2 Lite?
Nova 2 Lite is the stronger model overall, scoring 39.7 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Nova 2 Lite better for coding?
Nova 2 Lite scores higher on coding benchmarks: 40.7 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Nova 2 Lite share?
17 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Nova 2 Lite has 19.