Model comparison
DeepSeek V4 Flash vs Nova 2 Lite
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 39.7 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek V4 Flash scores higher in 8 categories and Nova 2 Lite in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Flash leads 53.7 to 27.5.
- DeepSeek V4 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.30 / $2.50 for Nova 2 Lite.
- DeepSeek V4 Flash has downloadable open weights; the other is API-only.
Side by side
| DeepSeek V4 Flash | Nova 2 Lite | |
|---|---|---|
| Provider | DeepSeek | Amazon |
| Noometry Index | 53.6 | 39.7 |
| Released | 2026-04-24 | 2025-12-01 |
| Weights | Open | Proprietary |
| Context window | 1M | 1M |
| Max output | 393K | 64K |
| Input $ / M tokens | $0.15 | $0.30 |
| Output $ / M tokens | $0.60 | $2.50 |
| Results tracked | 41 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Flash leads
DeepSeek V4 Flash: 47.9 (#59), Nova 2 Lite: 40.7 (#134)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1457 | 1385 |
| FrontierCode | 18.8% | — |
| LMArena WebDev | 1582 | — |
| SciCode | 49.9% | — |
| WeirdML | 63% | — |
| ALE-Bench | 1,306 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Flash: —, Nova 2 Lite: 24.1 (#121)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 27.1% |
Reasoning DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.7 (#30), Nova 2 Lite: 27.5 (#118)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1444 | 1364 |
| ARC-AGI-2 | 61.4% | — |
| SimpleBench | 61.1% | — |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 89.6% | — |
| ARC-AGI-1 | 89% | — |
| CritPt | 16.6% | — |
| Chess Puzzles | 33% | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 90.9% | — |
| LMCA | 41.7% | — |
| Epoch Capabilities Index | 154.49 | — |
Math DeepSeek V4 Flash leads
DeepSeek V4 Flash: 60.3 (#37), Nova 2 Lite: 37.5 (#156)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1427 | 1359 |
| FrontierMath (Tiers 1-3) | 57.5% | — |
| FrontierMath Tier 4 | 24.4% | — |
| MathArena Final-Answer Competitions | 76.5% | — |
| OTIS Mock AIME 2024-2025 | 94.4% | — |
| ProofBench | 56% | — |
Knowledge DeepSeek V4 Flash leads
DeepSeek V4 Flash: 55.4 (#48), Nova 2 Lite: 43.0 (#94)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Expert | 1441 | 1358 |
| GPQA Diamond | 91% | — |
| SimpleQA Verified | 33.6% | — |
| Vectara Hallucination Rate | — | 5.1% |
Multilingual DeepSeek V4 Flash leads
DeepSeek V4 Flash: 53.0 (#72), Nova 2 Lite: 47.1 (#153)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1420 | 1337 |
| LMArena Chinese | 1468 | 1364 |
| LMArena French | 1439 | 1381 |
| LMArena German | 1418 | 1343 |
| LMArena Japanese | 1406 | 1271 |
| LMArena Korean | 1384 | 1284 |
| LMArena Russian | 1428 | 1343 |
| LMArena Spanish | 1436 | 1373 |
Instruction Following DeepSeek V4 Flash leads
DeepSeek V4 Flash: 74.9 (#81), Nova 2 Lite: 70.5 (#161)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1421 | 1335 |
Long Context DeepSeek V4 Flash leads
DeepSeek V4 Flash: 43.8 (#85), Nova 2 Lite: 40.6 (#150)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1434 | 1335 |
Writing & Preference DeepSeek V4 Flash leads
DeepSeek V4 Flash: 63.8 (#61), Nova 2 Lite: 53.9 (#154)
| Benchmark | DeepSeek V4 Flash | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1432 | 1362 |
| LMArena Creative Writing | 1403 | 1291 |
| LMArena Multi-Turn | 1449 | 1338 |
| EQ-Bench Creative Writing | 1559 | — |
Frequently asked questions
Is DeepSeek V4 Flash better than Nova 2 Lite?
DeepSeek V4 Flash is the stronger model overall, scoring 53.6 to 39.7 on the Noometry Index.
Which is cheaper, DeepSeek V4 Flash or Nova 2 Lite?
DeepSeek V4 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Nova 2 Lite lists at $0.30 and $2.50.
Is DeepSeek V4 Flash or Nova 2 Lite better for coding?
DeepSeek V4 Flash scores higher on coding benchmarks: 47.9 versus 40.7 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do DeepSeek V4 Flash and Nova 2 Lite share?
17 benchmarks have published results for both models. DeepSeek V4 Flash has 41 scored results on Noometry and Nova 2 Lite has 19.