Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Gemini 4 Argon
Gemini 4 Argon has enough public results to be ranked (#24); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
DeepSeek
GoogleSummary
- The widest gap is in coding, where Gemini 4 Argon leads 62.0 to 40.4.
- DeepSeek-V2 (MoE-236B, May 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 40.3 | 56.5 |
| Released | 2024-05-07 | 2026-09-30 |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 10 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 4 Argon leads
DeepSeek-V2 (MoE-236B, May 2024): 40.4 (#139), Gemini 4 Argon: 62.0 (#9)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena WebDev | — | 1678 |
| FrontierSWE | — | 55% |
| SciCode | — | 61.8% |
| BigCodeBench Instruct | 48.9% | — |
| LMArena Coding | — | 1550 |
| BigCodeBench Complete | 59.4% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 49.2 (#8)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| APEX-Agents | — | 82.2% |
| Vending-Bench 2 | — | 13,718 |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 44.2 (#47)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| CritPt | — | 27.1% |
| LMArena Hard Prompts | — | 1548 |
| BIG-Bench Hard | 78.8% | — |
| Epoch Capabilities Index | 124.77 | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 73.9 (#17)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| ProofBench | — | 99% |
| LMArena Math | — | 1536 |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 43.9 (#89)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena Expert | — | 1553 |
| ARC (AI2) Challenge | 92.2% | — |
| MMLU | 78.4% | — |
| TriviaQA | 80% | — |
Multimodal Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 46.5 (#14)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| Blueprint-Bench 2 | — | 54.4% |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 60.3 (#1)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena Non-English | — | 1523 |
| LMArena Chinese | — | 1610 |
| LMArena Russian | — | 1537 |
| LMArena Spanish | — | 1497 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 80.1 (#2)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena Instruction Following | — | 1538 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 47.6 (#15)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena Longer Query | — | 1549 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Gemini 4 Argon: 71.4 (#19)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Gemini 4 Argon |
|---|---|---|
| LMArena Text | — | 1534 |
| LMArena Creative Writing | — | 1531 |
| LMArena Multi-Turn | — | 1557 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Gemini 4 Argon?
Gemini 4 Argon has enough public results to be ranked (#24); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Gemini 4 Argon better for coding?
Gemini 4 Argon scores higher on coding benchmarks: 62.0 versus 40.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Gemini 4 Argon share?
0 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Gemini 4 Argon has 21.