# DeepSeek-V3.1-Terminus vs Nova 2 Lite

> DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 39.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-terminus-vs-nova-2-lite
- Last updated: 2026-10-10
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 6 categories and Nova 2 Lite in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 54.1.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $0.30 / $2.50 for Nova 2 Lite.
- Nova 2 Lite accepts more context: 1M tokens versus 164K.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| Provider | DeepSeek | Amazon |
| Noometry Index | 42.6 | 39.0 |
| Rank | 96 | 161 |
| Context | 164K | 1M |
| Input $/M | $0.30 | $0.30 |
| Output $/M | $1 | $2.50 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek-V3.1-Terminus: 41.5 (#114)
- Nova 2 Lite: 40.1 (#131)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1426 | 1386 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |

## Agentic & Tool Use

- DeepSeek-V3.1-Terminus: —
- Nova 2 Lite: 23.9 (#114)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |

## Reasoning

- DeepSeek-V3.1-Terminus: 25.7 (#126)
- Nova 2 Lite: 26.4 (#120)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1427 | 1364 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |

## Math

- DeepSeek-V3.1-Terminus: 35.7 (#138)
- Nova 2 Lite: 34.8 (#154)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1402 | 1359 |

## Knowledge

- DeepSeek-V3.1-Terminus: —
- Nova 2 Lite: 42.5 (#88)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| Vectara Hallucination Rate | — | 5.1% |
| LMArena Expert | — | 1360 |

## Multilingual

- DeepSeek-V3.1-Terminus: 52.3 (#87)
- Nova 2 Lite: 47.3 (#148)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1407 | 1337 |
| LMArena Russian | 1434 | 1343 |
| LMArena Chinese | — | 1364 |
| LMArena French | — | 1381 |
| LMArena German | — | 1343 |
| LMArena Japanese | — | 1273 |
| LMArena Korean | — | 1284 |
| LMArena Spanish | — | 1372 |

## Instruction Following

- DeepSeek-V3.1-Terminus: 74.4 (#96)
- Nova 2 Lite: 70.7 (#156)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1406 | 1335 |

## Long Context

- DeepSeek-V3.1-Terminus: 43.5 (#88)
- Nova 2 Lite: 40.6 (#145)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1426 | 1335 |

## Writing & Preference

- DeepSeek-V3.1-Terminus: 61.3 (#87)
- Nova 2 Lite: 54.1 (#150)

| Benchmark | DeepSeek-V3.1-Terminus | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1420 | 1362 |
| LMArena Creative Writing | 1404 | 1291 |
| LMArena Multi-Turn | 1411 | 1338 |

## FAQ

### Is DeepSeek-V3.1-Terminus better than Nova 2 Lite?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 39.0 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.1-Terminus or Nova 2 Lite?

DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; Nova 2 Lite lists at $0.30 and $2.50.

### Is DeepSeek-V3.1-Terminus or Nova 2 Lite better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

Nova 2 Lite does, with 1M tokens against 164K.

### How many benchmarks do DeepSeek-V3.1-Terminus and Nova 2 Lite share?

10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Nova 2 Lite has 19.