All Releases•

41 models

DeepSeek Models: Intelligence, Performance & Price

Artificial Analysis has benchmarked 41 models from DeepSeek. Below is a comparison of the key metrics across these models.

  • For intelligence, the top model from DeepSeek is DeepSeek V4.1 Flash 1017 (Max) at 33.
  • For speed, DeepSeek V4.1 Flash 1017 (Non-reasoning) at 7.0 min has the lowest time per task.
  • For latency, DeepSeek V4.1 Flash 1017 (Non-reasoning) at 1.26s offers the lowest time to first answer token.
  • For pricing, DeepSeek V4.1 Flash 1017 (Non-reasoning) at $0.43 offers the lowest cost per task. Prices vary up to 4.1x across models.

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

Intelligence Index vs. Cost per Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task
Most attractive quadrant
Pareto line

Cost

Cost per Intelligence Index Task

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Performance

Time per Intelligence Index Task

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Speed at the model's own endpoint or the median across providers · Lower is better

Capability Scores

Capability Indexes

Measures the performance of models on specific capabilities and industries
Finance & Accounting Index

Incorporates 5 evaluations: AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf · Higher is better

Strategy & Ops Index

Incorporates 4 evaluations: AutomationBench-AA, AA-Briefcase v1.1, AA-Omniscience v1.1, GDP.pdf · Higher is better

Legal Index

Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf, AutomationBench-AA · Higher is better

Healthcare & Medical Index

Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, MLCR-AA, AutomationBench-AA, HLE-Diamond · Higher is better

Engineering Index

Incorporates 7 evaluations: AA-Omniscience v1.1, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1, HLE-Diamond, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Briefcase v1.1 · Higher is better

Economics Index

Incorporates 4 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf · Higher is better

All DeepSeek Releases

Further details

Weights
DeepSeek V4.1 Flash 1017 (Max)
DeepSeek logoDeepSeek
33
552B
16B active at inference time
1M
$0.18
13.1
177
BasetenNovitaGMI
+18
DeepSeek V4.1 Flash 1017 (High)
DeepSeek logoDeepSeek
29
552B
16B active at inference time
1M
$0.18
10.3
179
DeepSeek
DeepSeek V4 Pro 0813 (Max)
DeepSeek logoDeepSeek
23
1.6T
49B active at inference time
1M
$0.69
17.3
95
GMIDeepSeekSiliconFlow
+5
DeepSeek V4 Pro 0813 (High)
DeepSeek logoDeepSeek
21
1.6T
49B active at inference time
1M
$0.69
-
-
DeepSeek
DeepSeek V4.1 Flash 1017 (Non-reasoning)
DeepSeek logoDeepSeek
19
552B
16B active at inference time
1M
$0.18
7.0
155
DeepSeekIncoModular
+3
DeepSeek V4 Pro 0813 (Non-reasoning)
DeepSeek logoDeepSeek
13
1.6T
49B active at inference time
1M
$0.69
13.8
87
DeepSeekNebiusDigitalOcean
DeepSeek V3.2 Exp (Non-reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.12
-
-
DeepSeek
DeepSeek V3.1 Terminus (Reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$1.70
-
-
NovitaSambaNova
DeepSeek R1 (Jan '25)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$2.20
-
129
Amazon BedrockNovitaTogether AIHyperbolic
DeepSeek R1 Distill Llama 8B
DeepSeek logoDeepSeek
-
8B
128k
-
-
-
-
DeepSeek V3.1 (Reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.47
-
107
Amazon BedrockNovita
DeepSeek-V2-Chat
DeepSeek logoDeepSeek
-
236B
21B active at inference time
128k
-
-
-
-
DeepSeek V3 (Dec '24)
DeepSeek logoDeepSeek
-
671B
37B active at inference time
128k
$0.38
-
77
HyperbolicDeepInfraNovita
DeepSeek R1 Distill Qwen 1.5B
DeepSeek logoDeepSeek
-
1.5B
128k
-
-
-
-
DeepSeek R1 Distill Llama 70B
DeepSeek logoDeepSeek
-
70B
128k
$0.74
-
-
DeepInfraSambaNova
DeepSeek V3.2 Speciale
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
-
-
-
-
DeepSeek V4 Pro 0424 (High)
DeepSeek logoDeepSeek
-
1.6T
49B active at inference time
1M
$0.73
-
-
DeepInfraMicrosoft AzureSiliconFlow
+2
DeepSeek R1 0528 Qwen3 8B
DeepSeek logoDeepSeek
-
8.2B
33k
-
-
-
-
DeepSeek V3.2 (Non-reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.29
-
42
GMIFireworksMicrosoft Azure
+5
DeepSeek V4 Pro 0424 (Non-reasoning)
DeepSeek logoDeepSeek
-
1.6T
49B active at inference time
1M
$0.80
-
136
Microsoft AzureCoreWeaveBaseten
DeepSeek V4.1 Flash (Non-reasoning)
DeepSeek logoDeepSeek
-
552B
16B active at inference time
1M
$0.18
-
221
ModalFireworksCoreWeave
+5
DeepSeek V3 0324
DeepSeek logoDeepSeek
-
671B
37B active at inference time
128k
$0.23
-
139
DeepInfra
DeepSeek V4 Flash Vision (Max)
DeepSeek logoDeepSeek
-
284B
13B active at inference time
1M
$0.23
-
167
Not available
DeepInfraGMISiliconFlow
DeepSeek LLM 67B Chat (V1)
DeepSeek logoDeepSeek
-
7B
4k
-
-
-
-
DeepSeek V4 Flash 0420 (High)
DeepSeek logoDeepSeek
-
284B
13B active at inference time
1M
$0.0622
-
-
DeepInfraGMIParasailSiliconFlow
DeepSeek-V2.5
DeepSeek logoDeepSeek
-
236B
21B active at inference time
128k
-
-
-
-
DeepSeek-Coder-V2
DeepSeek logoDeepSeek
-
236B
21B active at inference time
128k
-
-
-
-
DeepSeek R1 0528 (May '25)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$1.82
-
33
Together AIHyperbolicNovitaDeepInfra
DeepSeek V4 Flash 0420 (Non-reasoning)
DeepSeek logoDeepSeek
-
284B
13B active at inference time
1M
$0.0508
-
-
GMI
DeepSeek V3.2 Exp (Reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.12
-
-
DeepSeek
DeepSeek R1 Distill Qwen 14B
DeepSeek logoDeepSeek
-
14B
128k
-
-
-
-
DeepSeek-V2.5 (Dec '24)
DeepSeek logoDeepSeek
-
236B
21B active at inference time
128k
-
-
-
-
DeepSeek V4 Pro 0424 (Max)
DeepSeek logoDeepSeek
-
1.6T
49B active at inference time
1M
$0.71
-
82
DeepInfraNovitaSiliconFlow
+4
DeepSeek Coder V2 Lite Instruct
DeepSeek logoDeepSeek
-
16B
2.4B active at inference time
128k
-
-
-
-
DeepSeek V3.2 (Reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.11
-
40
DeepInfraAmazon BedrockFireworks
+3
DeepSeek V4 Flash 0420 (Max)
DeepSeek logoDeepSeek
-
284B
13B active at inference time
1M
$0.0622
-
125
GMIDeepInfraSiliconFlowParasail
DeepSeek V4.1 Flash (Max)
DeepSeek logoDeepSeek
-
552B
16B active at inference time
1M
$0.18
-
242
CoreWeaveModalSiliconFlow
+14
DeepSeek R1 Distill Qwen 32B
DeepSeek logoDeepSeek
-
32B
128k
-
-
-
-
DeepSeek V3.1 Terminus (Non-reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.25
-
101
DeepInfraSambaNovaNovita
DeepSeek V3.1 (Non-reasoning)
DeepSeek logoDeepSeek
-
685B
37B active at inference time
128k
$0.53
-
65
Amazon BedrockCoreWeaveDeepInfra
+3
DeepSeek V4 Flash 0731 (Max)
DeepSeek logoDeepSeek
-
284B
13B active at inference time
1M
$0.0756
-
148
CoreWeaveWaferSiliconFlow
+7