# DeepSeek-V3.1-Terminus vs DeepSeek V4.1 Flash

> DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 43.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-terminus-vs-deepseek-v4-1-flash
- Last updated: 2026-10-12
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and DeepSeek V4.1 Flash in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 38.5.
- The biggest single-benchmark swing is LMCA: 28.6% for DeepSeek-V3.1-Terminus and 47% for DeepSeek V4.1 Flash.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.27 / $1 for DeepSeek-V3.1-Terminus.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 164K.

## Snapshot

| | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 43.1 | 52.8 |
| Rank | 97 | 38 |
| Context | 164K | 1M |
| Input $/M | $0.27 | $0.15 |
| Output $/M | $1 | $0.60 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1-Terminus: 42.0 (#113)
- DeepSeek V4.1 Flash: 52.9 (#32)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| SciCode | 40.6% | 51.9% |
| LMArena Coding | 1426 | 1506 |
| ALE-Bench | 745.17 | 1,092 |
| LMArena WebDev | — | 1619 |

## Agentic & Tool Use

- DeepSeek-V3.1-Terminus: —
- DeepSeek V4.1 Flash: 31.2 (#69)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| APEX-Agents | — | 39.5% |
| GDP.pdf | — | 19.8% |

## Reasoning

- DeepSeek-V3.1-Terminus: 26.4 (#133)
- DeepSeek V4.1 Flash: 50.2 (#36)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| CritPt | 1.7% | 14.3% |
| LMArena Hard Prompts | 1426 | 1483 |
| DTBench | 81.3% | 89.9% |
| LMCA | 28.6% | 47% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 89.6% |
| Mystery Game Puzzles | — | 43% |
| Surface Evolver Bench | — | 46.3% |
| Epoch Capabilities Index | — | 154.9 |

## Math

- DeepSeek-V3.1-Terminus: 38.5 (#137)
- DeepSeek V4.1 Flash: 66.7 (#25)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Math | 1402 | 1477 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 26.8% |
| OTIS Mock AIME 2024-2025 | — | 98.3% |
| ProofBench | — | 54% |

## Knowledge

- DeepSeek-V3.1-Terminus: —
- DeepSeek V4.1 Flash: 57.9 (#38)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| GPQA Diamond | — | 89.8% |
| LMArena Expert | — | 1506 |

## Multimodal

- DeepSeek-V3.1-Terminus: —
- DeepSeek V4.1 Flash: 39.1 (#61)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Vision | — | 1277 |
| Furniture Assembly | — | 34.2% |

## Multilingual

- DeepSeek-V3.1-Terminus: 52.1 (#92)
- DeepSeek V4.1 Flash: 55.0 (#35)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Non-English | 1407 | 1448 |
| LMArena Russian | 1436 | 1471 |
| LMArena Chinese | — | 1497 |
| LMArena French | — | 1452 |
| LMArena German | — | 1484 |
| LMArena Japanese | — | 1412 |
| LMArena Korean | — | 1452 |
| LMArena Spanish | — | 1459 |

## Instruction Following

- DeepSeek-V3.1-Terminus: 74.0 (#106)
- DeepSeek V4.1 Flash: 77.3 (#26)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Instruction Following | 1404 | 1474 |

## Long Context

- DeepSeek-V3.1-Terminus: 43.4 (#97)
- DeepSeek V4.1 Flash: 45.2 (#47)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Longer Query | 1421 | 1475 |

## Writing & Preference

- DeepSeek-V3.1-Terminus: 61.0 (#92)
- DeepSeek V4.1 Flash: 65.4 (#48)

| Benchmark | DeepSeek-V3.1-Terminus | DeepSeek V4.1 Flash |
|---|---|---|
| LMArena Text | 1419 | 1462 |
| LMArena Creative Writing | 1403 | 1435 |
| LMArena Multi-Turn | 1411 | 1457 |
| EQ-Bench Creative Writing | — | 1540 |

## FAQ

### Is DeepSeek-V3.1-Terminus better than DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 43.1 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.1-Terminus or DeepSeek V4.1 Flash?

DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; DeepSeek-V3.1-Terminus lists at $0.27 and $1.

### Is DeepSeek-V3.1-Terminus or DeepSeek V4.1 Flash better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 42.0 in the Noometry coding category.

### Which has the bigger context window?

DeepSeek V4.1 Flash does, with 1M tokens against 164K.

### How many benchmarks do DeepSeek-V3.1-Terminus and DeepSeek V4.1 Flash share?

15 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and DeepSeek V4.1 Flash has 37.