DeepSeek benchmarks.
A lab-level profile for DeepSeek, covering tracked models, category leaders, open/proprietary mix, speed, value, source coverage, and portfolio depth.
Lab rank
#16
By available Arena average
Models tracked
9
9 open-weight models
Portfolio coverage
22%
Average matched benchmark signals
Fastest model
0.66s
DeepSeek V4.1 Flash
Lab profile
Portfolio shape, leaders, and model coverage.
The lab page now starts with the portfolio read, then drills into category leaders, value rankings, and model-level coverage rather than repeating source panels.
Portfolio read
DeepSeek has 9 tracked benchmark profiles.
The portfolio combines 0 proprietary models and 9 open-weight models. Its average available Arena rank is 59.5, with a best category rank of #24. Average profile coverage is 22%, so model-level gaps remain visible in the cards below.
Best Arena model
DeepSeek V4.1 Flash
Best category rank #24
Highest coverage
DeepSeek V4 Pro
56% matched signals
Fastest response
DeepSeek V4.1 Flash
0.66s first token
Access mix
Open versus proprietary
Proprietary
0 / 0%
Open weights
9 / 100%
Value leader
DeepSeek V4.1 FlashCategory leaders
Best model by benchmark lens.
Each card shows the highest-ranked model from this lab in a composite benchmark suite, with the global rank and index left intact.
Value
Quality, price, latency, and speed
DeepSeek V4.1 Flash
value index
Coding
Code Arena, SWE-bench, and output speed
DeepSeek V4.1 Flash
coding index
Agents
Search, document, code, and latency
DeepSeek V4.1 Flash
agent proxy
Reasoning
Text, document, vision, and mode signals
DeepSeek V4.1 Flash Max
reasoning proxy
Context
Document, search, and text proxy
DeepSeek V4.1 Flash Max
context proxy
DeepSeek value ranking table
Models from this lab with both quality and commercial price matches.
Top 3 of 3
| Rank | Model | Index | Price | Latency | Speed | Quality | Sources |
|---|---|---|---|---|---|---|---|
| #10 | 75value index | $0.525blended / 1M tokens | 0.66stime to first token | 351 tok/soutput speed | 55quality signal | ArenaArtificial Analysis | |
| #29 | 65value index | $0.175blended / 1M tokens | Not listedtime to first token | Not listedoutput speed | 46quality signal | ArenaArtificial Analysis | |
| #75 | 48value index | $1.98blended / 1M tokens | 1.03stime to first token | 93 tok/soutput speed | 19quality signal | ArenaArtificial Analysis |
Model portfolio
DeepSeek model coverage map.
Every tracked model keeps its best Arena rank, coverage score, commercial metrics, and suite placement visible in one scannable grid.
DeepSeek V4.1 Flash
DeepSeek-V4.1-Flash (Max)
Coverage
44%
Best Arena
#24
Code
Value
#10
value index
Coding
#10
coding index
Latency
0.66s
351 tok/s
DeepSeek V4 Flash
DeepSeek-V4-Flash (High)
Coverage
22%
Best Arena
#34
Code
Value
#29
value index
Coding
#26
coding index
Latency
Not listed
Not listed
DeepSeek V4.1 Flash Max
deepseek-v4.1-flash-max
Coverage
22%
Best Arena
#38
Text
Value
Not listed
No price match
Coding
Not listed
No coding index
Latency
Not listed
Not listed
DeepSeek V4 Pro High 20260813
deepseek-v4-pro-high-20260813
Coverage
11%
Best Arena
#56
Text
Value
Not listed
No price match
Coding
Not listed
No coding index
Latency
Not listed
Not listed
DeepSeek V4 Pro
deepseek-v4-pro
Coverage
56%
Best Arena
#64
Text
Value
#75
value index
Coding
#64
coding index
Latency
1.03s
93 tok/s
DeepSeek V4 Pro Preview
DeepSeek-V4-Pro Preview (High)
Coverage
11%
Best Arena
#68
Code
Value
Not listed
No price match
Coding
#59
coding index
Latency
Not listed
Not listed
DeepSeek V4 Pro High Preview
deepseek-v4-pro-high-preview
Coverage
11%
Best Arena
#71
Text
Value
Not listed
No price match
Coding
Not listed
No coding index
Latency
Not listed
Not listed
DeepSeek V4 Flash Preview
DeepSeek-V4-Flash Preview (High)
Coverage
11%
Best Arena
#84
Code
Value
Not listed
No price match
Coding
#75
coding index
Latency
Not listed
Not listed
DeepSeek V4 Flash High Preview
deepseek-v4-flash-high-preview
Coverage
11%
Best Arena
#98
Text
Value
Not listed
No price match
Coding
Not listed
No coding index
Latency
Not listed
Not listed
