Model comparison
Deepseek Coder v2 vs Nova Premier 1.0
Nova Premier 1.0 is the stronger model overall, scoring 38.3 to 35.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
DeepSeekSummary
- The widest gap is in writing & preference, where Nova Premier 1.0 leads 51.0 to 38.2.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | Nova Premier 1.0 | |
|---|---|---|
| Provider | DeepSeek | Amazon |
| Noometry Index | 35.9 | 38.3 |
| Released | 2024-06-17 | 2025-04-30 |
| Weights | Open | Proprietary |
| Context window | — | 1M |
| Max output | — | 10K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $12.50 |
| Results tracked | 24 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Deepseek Coder v2: 38.1 (#183), Nova Premier 1.0: —
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| BigCodeBench Instruct | 48.2% | — |
| LMArena Coding | 1251 | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Reasoning Too close to call
Deepseek Coder v2: 23.6 (#176), Nova Premier 1.0: 23.0 (#185)
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| Kagi LLM Benchmark | — | 44.8% |
| LMArena Hard Prompts | 1207 | — |
| WinoGrande | 83.7% | — |
Math Deepseek Coder v2 leads
Deepseek Coder v2: 34.9 (#190), Nova Premier 1.0: 33.8 (#199)
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| Omni-MATH | — | 35% |
| LMArena Math | 1241 | — |
| GSM8K | 94.5% | — |
Knowledge Nova Premier 1.0 leads
Deepseek Coder v2: 32.3 (#212), Nova Premier 1.0: 35.8 (#180)
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| MMLU-Pro | — | 72.6% |
| GPQA (HELM) | — | 51.8% |
| LMArena Expert | 1181 | — |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual Not comparable
Deepseek Coder v2: 36.3 (#240), Nova Premier 1.0: —
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| LMArena Non-English | 1182 | — |
| LMArena Chinese | 1201 | — |
| LMArena French | 1185 | — |
| LMArena German | 1164 | — |
| LMArena Japanese | 1126 | — |
| LMArena Korean | 1104 | — |
| LMArena Russian | 1188 | — |
| LMArena Spanish | 1153 | — |
Instruction Following Nova Premier 1.0 leads
Deepseek Coder v2: 61.7 (#242), Nova Premier 1.0: 66.8 (#204)
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| IFEval | — | 80.3% |
| LMArena Instruction Following | 1180 | — |
Long Context Not comparable
Deepseek Coder v2: 37.0 (#224), Nova Premier 1.0: —
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| LMArena Longer Query | 1219 | — |
Writing & Preference Nova Premier 1.0 leads
Deepseek Coder v2: 38.2 (#253), Nova Premier 1.0: 51.0 (#178)
| Benchmark | Deepseek Coder v2 | Nova Premier 1.0 |
|---|---|---|
| LMArena Text | 1191 | — |
| LMArena Creative Writing | 1120 | — |
| WildBench | — | 78.8% |
| LMArena Multi-Turn | 1177 | — |
Frequently asked questions
Is Deepseek Coder v2 better than Nova Premier 1.0?
Nova Premier 1.0 is the stronger model overall, scoring 38.3 to 35.9 on the Noometry Index.
How many benchmarks do Deepseek Coder v2 and Nova Premier 1.0 share?
0 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and Nova Premier 1.0 has 6.