DeepSeek V4 Flash
| Released | 2026-04-24 |
|---|---|
| Price per 1M | $0.14 in · $0.28 out |
| Overall score | 53.6 · rank 85 |
| Context window | 1,000,000 tokens |
| Provider | DeepSeek |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 53.6 | 85 | 10 |
| Overall: Coding | 52.3 | 89 | 6 |
| Overall: Reasoning | 55.8 | 68 | 5 |
| Overall: Agentic work | 52.7 | 95 | 5 |
| Overall: Frontend | 54.0 | 66 | 2 |
| Coding | 50.7 | 112 | 1 |
| Reasoning | 55.9 | 78 | 4 |
| Math | 55.7 | 63 | 1 |
| Frontend | 55.6 | 50 | 1 |
| General | 54.0 | 74 | 3 |
| Value vs price | +11.0 | 26 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 53.8 ±6.5 | 6 |
| High | 50.1 ±4.7 | 3 |
| Max | 53.3 ±4.3 | 3 |
Evidence
Coding 50.7 ±4.7
Reasoning 55.9 ±5.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Extended NYT Connections | 89.6% | Lech Mazur | 2026-11-10 |
| CritPt | 7.1% | Epoch AI | 2026-11-11 |
| Kagi Reasoning Benchmark | 52.2% | Kagi | 2026-11-11 |
| SimpleBench | 61.1% | SimpleBench | 2026-08-03 |
| SimpleBench | 46.3% | Epoch AI | 2026-11-11 |
Math 55.7 ±4.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MathArena Final-Answer Competitions | 76.6% | MathArena | 2026-11-11 |
Frontend 55.6 ±3.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LMArena WebDev | 1,582 | LMArena | 2026-11-09 |
General 54.0 ±3.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| EQ-Bench Creative Writing v3 | 1,559 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,436 | LMArena | 2026-11-09 |
| EQ-Bench Longform Creative Writing | 66.0% | EQ-Bench | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| CritPt | 3.4% | Epoch AI | A stronger configuration result was used |
| SciCode | 42.0% | Epoch AI | A stronger configuration result was used |
Names matched to this model
deepseek-v4-flash