# DeepSeek LLM 67B vs Nova 2 Lite

> Nova 2 Lite is the stronger model overall, scoring 39.7 to 24.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-nova-2-lite
- Last updated: 2026-10-11
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and Nova 2 Lite in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nova 2 Lite leads 43.0 to 7.0.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| Provider | DeepSeek | Amazon |
| Noometry Index | 24.9 | 39.7 |
| Rank | 347 | 161 |
| Context | — | 1M |
| Input $/M | — | $0.30 |
| Output $/M | — | $2.50 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- Nova 2 Lite: 40.7 (#134)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1096 | 1385 |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- Nova 2 Lite: 24.1 (#121)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- Nova 2 Lite: 27.5 (#118)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1364 |
| Chess Puzzles | 0% | — |
| Epoch Capabilities Index | 110.5 | — |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- Nova 2 Lite: 37.5 (#156)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1108 | 1359 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| MATH Level 5 | 6.4% | — |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- Nova 2 Lite: 43.0 (#94)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| Vectara Hallucination Rate | — | 5.1% |
| LMArena Expert | — | 1358 |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- Nova 2 Lite: 47.1 (#153)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1073 | 1337 |
| LMArena Chinese | 1132 | 1364 |
| LMArena French | — | 1381 |
| LMArena German | — | 1343 |
| LMArena Japanese | — | 1271 |
| LMArena Korean | — | 1284 |
| LMArena Russian | — | 1343 |
| LMArena Spanish | — | 1373 |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- Nova 2 Lite: 70.5 (#161)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1079 | 1335 |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- Nova 2 Lite: 40.6 (#150)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1092 | 1335 |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- Nova 2 Lite: 53.9 (#154)

| Benchmark | DeepSeek LLM 67B | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1105 | 1362 |
| LMArena Creative Writing | 1067 | 1291 |
| LMArena Multi-Turn | 1082 | 1338 |

## FAQ

### Is DeepSeek LLM 67B better than Nova 2 Lite?

Nova 2 Lite is the stronger model overall, scoring 39.7 to 24.9 on the Noometry Index.

### Is DeepSeek LLM 67B or Nova 2 Lite better for coding?

Nova 2 Lite scores higher on coding benchmarks: 40.7 versus 31.9 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and Nova 2 Lite share?

10 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Nova 2 Lite has 19.