Data points: 95
Model compare
The readout for DeepSeek V3 and DeepSeek V3.1, before the detailed comparison sheet.
Weighted outcome: DeepSeek V3.1. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
Decision read
DeepSeek V3.1
DeepSeek V3.1 has the higher weighted result; Model A / B score 25.3 to 74.7.
Evidence depth
95 data points
Includes 24 benchmark rows, 0 audit samples, and 7 provider examples.
Selection signal
Start with DeepSeek V3.1
The charts below split 31 high-signal samples across speed, scores, and audit health.
Switch either side of this report to compare another model with the same LMSpeed data pipeline.
Select a different model to open a new comparison URL.
This report only uses LMSpeed data for DeepSeek V3 and DeepSeek V3.1: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
| Model compare | DeepSeek V3 | DeepSeek V3.1 |
|---|---|---|
| Overall leader | Contender | Leading |
| Weighted overall score | 25.3 pts | 74.7 pts |
| Benchmark category leads | 1 categories | 5 categories |
| Operational advantages | Cheapest input price, Free providers, Provider coverage | Average speed, First-token latency |
| Context window | No data | No data |
| Max output | No data | No data |
| Modalities | No data | No data |
| Features | None listed | None listed |
The overall result weights benchmark capability categories at 80% and price, API speed/latency, and availability at 20%. Recent test volume does not affect the winner, and missing benchmark categories are excluded.
| Model compare | DeepSeek V3 | DeepSeek V3.1 |
|---|---|---|
| Developer | No data | DeepSeek |
| Released | No data | No data |
| Parameters | No data | No data |
| Tokenizer | No data | No data |
| Knowledge cutoff | No data | No data |
| OpenRouter ID | No data | No data |
| References | No data | No data |
This report only uses LMSpeed data for DeepSeek V3 and DeepSeek V3.1: pricing, speed aggregates, third-party benchmark scores, and shared provider samples.
DeepSeek V3
DeepSeek V3 is stronger in benchmark categories (Knowledge) and operational dimensions (Cheapest input price, Free providers, Provider coverage).
DeepSeek V3.1
DeepSeek V3.1 is stronger in benchmark categories (Agents, Coding, Reasoning, Math, Instruction following) and operational dimensions (Average speed, First-token latency).
Third-party benchmark profile synced into LMSpeed; only metrics available for both models are shown.
Compare benchmark category scores on a 0-100 scale. Select a category to inspect the gap.
Avg. score
DeepSeek V3
41.7
Avg. score
DeepSeek V3.1
43.2
DeepSeek V3.1 leads by 4.9
DeepSeek V3.1 leads by 15.1
DeepSeek V3.1 leads by 5.2
DeepSeek V3 leads by 8.1
DeepSeek V3.1 leads by 3.5
DeepSeek V3.1
DeepSeek V3.1
DeepSeek V3.1 leads by 3.0
Metric-level scores with benchmark source, rank depth, confidence, error, and evaluation date where available.
DeepSeek V3
winner38.0
Rank #89/127 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
31.0
Rank #99/127 · confidence 1 · eval date 2025-08-21
DeepSeek V3
winner$0.890/M
Rank #38/163 · confidence 4
DeepSeek V3.1
$1.68/M
Rank #59/163 · confidence 4
DeepSeek V3
winner$0.493/M
Rank #40/163 · confidence 4
DeepSeek V3.1
$0.840/M
Rank #67/163 · confidence 4
DeepSeek V3
winner$0.360/M
Rank #60/163 · confidence 4
DeepSeek V3.1
$0.560/M
Rank #79/163 · confidence 4
DeepSeek V3
75.2%
Rank #85/125 · confidence 4
DeepSeek V3.1
winner83.3%
Rank #39/125 · confidence 4
DeepSeek V3
21.5
Rank #65/70 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner26.4
Rank #62/70 · confidence 1 · eval date 2025-08-21
DeepSeek V3
29.0
Rank #81/89 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner53.3
Rank #65/89 · confidence 1 · eval date 2025-08-21
DeepSeek V3
55.7%
Rank #154/187 · confidence 4
DeepSeek V3.1
winner73.5%
Rank #105/187 · confidence 4
DeepSeek V3
3.6%
Rank #172/186 · confidence 4
DeepSeek V3.1
winner6.3%
Rank #121/186 · confidence 4
DeepSeek V3
35.9%
Rank #78/115 · confidence 4
DeepSeek V3.1
winner57.7%
Rank #46/115 · confidence 4
DeepSeek V3
6.8
Rank #77/85 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner25.0
Rank #60/85 · confidence 1 · eval date 2025-08-21
DeepSeek V3
35.4
Rank #77/96 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner39.1
Rank #63/96 · confidence 1 · eval date 2025-08-21
DeepSeek V3
winner33.0
Rank #83/100 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
19.0
Rank #94/100 · confidence 1 · eval date 2025-08-21
DeepSeek V3
35.4%
Rank #109/184 · confidence 4
DeepSeek V3.1
winner36.7%
Rank #97/184 · confidence 4
DeepSeek V3
winner89.4
Rank #15/90 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
80.3
Rank #38/90 · confidence 1 · eval date 2025-08-21
DeepSeek V3
25.4
Rank #54/90 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner28.8
Rank #47/90 · confidence 1 · eval date 2025-08-21
DeepSeek V3
3.6
Rank #93/96 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner13.0
Rank #64/96 · confidence 1 · eval date 2025-08-21
DeepSeek V3
winner50.5
Rank #64/92 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
18.4
Rank #87/92 · confidence 1 · eval date 2025-08-21
DeepSeek V3
55.7
Rank #90/96 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner77.9
Rank #65/96 · confidence 1 · eval date 2025-08-21
DeepSeek V3
14.2
Rank #85/99 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner20.7
Rank #73/99 · confidence 1 · eval date 2025-08-21
DeepSeek V3
1154.0
Rank #52/65 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner1157.0
Rank #50/65 · confidence 1 · eval date 2025-08-21
DeepSeek V3
winner61.1
Rank #51/75 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
30.5
Rank #69/75 · confidence 1 · eval date 2025-08-21
DeepSeek V3
34.8
Rank #81/86 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner41.5
Rank #67/86 · confidence 1 · eval date 2025-08-21
DeepSeek V3
22.8
Rank #78/84 · confidence 1 · eval date 2024-12-26
DeepSeek V3.1
winner37.4
Rank #71/84 · confidence 1 · eval date 2025-08-21
Latest completed audits from shared providers, with four safety and integrity score groups plus report links.
| Provider | DeepSeek V3 | DeepSeek V3.1 |
|---|---|---|
| No completed audits are available from shared providers yet. | ||
Speed aggregates and input/output pricing share each provider row for real API selection and migration cost checks.
| Provider | DeepSeek V3 | DeepSeek V3.1 |
|---|---|---|
75 tests | DeepSeek V3 speed / latency 33 tok/s / 1330ms input / output No data | DeepSeek V3.1 speed / latency N/A / N/A input / output No data |
35 tests | DeepSeek V3 speed / latency N/A / N/A input / output No data | DeepSeek V3.1 speed / latency 265 tok/s / 951ms input / output No data |
35 tests | DeepSeek V3 speed / latency 109 tok/s / 5107ms input / output No data | DeepSeek V3.1 speed / latency 134 tok/s / 1688ms input / output No data |
31 tests | DeepSeek V3 speed / latency N/A / N/A input / output No data | DeepSeek V3.1 speed / latency 68 tok/s / 3024ms input / output No data |
30 tests | DeepSeek V3 speed / latency 95 tok/s / 519ms input / output No data | DeepSeek V3.1 speed / latency N/A / N/A input / output No data |
DeepSeek V3 deepseek-v3 speed / latency No data input / output $0/M/$0/M | DeepSeek V3.1 deepseek-v3.1 speed / latency No data input / output $0.038/M | |
DeepSeek V3 deepseek-v3-0324 speed / latency No data input / output $0/request | DeepSeek V3.1 deepseek-v3.1 speed / latency No data input / output $0/request |
Weighted outcome: DeepSeek V3.1. Benchmark capability categories carry 80%, while price, API performance, and availability carry 20%.
