modelbenchmark.io

DeepSeek-V3 (Deep Infra)

DeepSeek · deepseek

Compare

Open DeepSeek MoE chat model for coding, math, and general reasoning

Specification

deepseek listing

Context
66K
Max output
8K
Released
2024-12-26
Knowledge cutoff
2024-07
Retires
2027-11-01
Open weights
yes
Input
text
Output
text

Price

US dollars per million tokens · deepseek list price

Input
$0.27
Output
$1.10
Cache read
$0.07
Cache write
free

Quality

6 benchmarks · 3 sources

Composite
71st
percentile of 344 scored models
Rank
99 / 344
±0.06 sd
Evidence
6 × 3
3 or more benchmarks, from 2 or more sources
Effort range
10.0
points between effort settings
coding70th2/3 bench
reasoning78th1/4 bench
math55th3/7 bench

Every score

one row per source and configuration — nothing averaged away

BenchmarkScoreConfigurationSourceRun
Aider polyglot72.2default · aider diff · 2 runsAider2025-10-03
GPQA diamond77.3 ±3.2default · 2 runsEpoch AI2026-07-16
SWE-Bench verified70.0high · mini-SWE-agentSWE-bench—
↳60.0default · mini-SWE-agentSWE-bench—
Settings differ by 10.0 points on this benchmark.
OTIS Mock AIME 2024-202568.4 ±7.5default · 2 runsEpoch AI2026-07-16
FrontierMath-v122.1 ±4.1defaultEpoch AI2025-12-22
FrontierMath-Tier-4-2025-07-012.1 ±2.1defaultEpoch AI2025-12-16

The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.

Available from 28 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
deepseekvendor
$0.27
$0.28
$1.10
$0.40
$0.07
$0.028
free
66K
164K
8K
164K
—
tokengo
$0.19
$0.2174
$0.71
$0.326
$0.06—
131K
128K
8K
64K
—
hyperbolic
$0.20
$0.40
$0.20
$0.40
——33K33K—
lambda_ai$0.20$0.60——131K131K—
nano-gpt · thinking$0.2088$0.3096$0.0216—163K66K—
zenmux
$0.22
$0.293
$0.33
$0.4395
$0.0293—
163K
128K
64K
8K
—
vercel
$0.25
$0.62
$0.95
$1.85
$0.13—
164K
128K
128K
8K
—
llmgateway-providers$0.26$0.38$0.13—160K66K—
poe$0.2626$0.3838$0.1313—128K0—
novita
$0.269
$0.27
$0.89
$0.40
$0.41
$1.00
$1.12
$0.89
$0.1345
$0.135
—
164K
131K
64K
66K
33K
16K
—
kilo
$0.27
$0.26
$0.41
$0.38
$0.13—164K147K—
openrouter
$0.27
$0.26
$0.41
$0.38
$0.13—164K147K—
jiekou
$0.28
$0.27
$1.14
$1.00
$0.14—164K
164K
33K
—
nano-gpt
$0.2838
$0.2088
$0.8514
$0.3096
$0.009
$0.0216
—
1M
163K
384K
66K
—
qiniu-ai$0.2899$0.4348——128K32K—
ofox$0.29$0.43$0.06—128K32K—
tensorx$0.30$0.50$0.075$0.375164K164K—
deepinfra
$0.32
$0.24
$0.25
$0.26
$0.89
$0.90
$0.95
$0.38
$0.135
$0.216
$0.13
—164K164K—
edenai
$0.32
$0.24
$0.89
$0.90
$0.135—164K
8K
164K
—
baseten
$0.50
$0.77
$1.50
$0.77
————2026-10-16
nebius$0.50$1.50——128K128K—
abacus$0.55$1.66——128K8K—
merge-gateway
$0.56
$0.28
$0.27
$1.68
$0.40
$1.10
$0.13—
131K
128K
164K
33K
82K
—
azure_ai
$0.58
$1.14
$1.23
$1.68
$4.56
$4.94
——128K
128K
8K
2027-11-01
2025-08-31
2026-07-13
novita-ai$0.89$0.89——64K16K—
fireworks_ai
$0.90
$0.56
$1.20
$0.90
$1.68
$1.20
——
131K
164K
128K
33K
8K
164K
33K
—
nearai$1.10$1.10$0.55—128K64K—
together_ai
$1.25
$0.60
$1.25
$1.70
——
66K
131K
8K
131K
2026-05-14

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02deepseekCache read$0.07
2026-09-22deepseekCache read$0.07
2026-09-02kiloCache read$0.1345
2026-09-29kiloCache read$0.0672
2026-09-30kiloCache read$0.028
2026-10-08kiloCache read$0.135
2026-10-11kiloCache read$0.14
2026-10-18kiloCache read$0.028
2026-10-31kiloCache read$0.13
2026-09-02llmgateway-providersCache read$0.13
2026-09-04llmgateway-providersCache read$0.1345
2026-09-24llmgateway-providersCache read$0.13
2026-10-20llmgateway-providersCache read$0.1345
2026-10-21llmgateway-providersCache read$0.13
2026-10-24llmgateway-providersCache read$0.1345
2026-10-25llmgateway-providersCache read$0.13
2026-10-28llmgateway-providersCache read$0.1345
2026-10-29llmgateway-providersCache read$0.13
2026-09-02merge-gatewayCache read$0.14
2026-10-01merge-gatewayCache read$0.13
2026-09-02nano-gptCache read$0.14
2026-09-24nano-gptCache read$0.04
2026-10-13nano-gptCache read$0.10
2026-10-14nano-gptCache read$0.04
2026-10-15nano-gptCache read$0.1866
2026-10-16nano-gptCache read$0.04
2026-10-19nano-gptCache read$0.022
2026-10-21nano-gptCache read$0.0216
2026-10-24nano-gptCache read$0.0216
2026-10-26nano-gptCache read$0.0216
2026-10-27nano-gptCache read$0.022
2026-10-28nano-gptCache read$0.0216
2026-10-31nano-gptCache read$0.022
2026-11-01nano-gptCache read$0.009
2026-09-02nano-gpt · thinkingCache read$0.14
2026-10-16nano-gpt · thinkingCache read$0.0216
2026-09-02novita-aiCache read$0.135
2026-09-04novita-aiCache read$0.1345
2026-09-24novita-aiCache read$0.135
2026-09-02openrouterCache read$0.1345
2026-09-19openrouterCache read$0.1345
2026-09-22openrouterCache read$0.1345
2026-09-28openrouterCache read$0.028
2026-09-29openrouterCache read$0.0672
2026-09-30openrouterCache read$0.028
2026-10-08openrouterCache read$0.135
2026-10-09openrouterCache read$0.135
2026-10-10openrouterCache read$0.135
2026-10-11openrouterCache read$0.14
2026-10-18openrouterCache read$0.028
2026-10-31openrouterCache read$0.13
2026-09-02poeCache read$0.13
2026-10-22poeCache read$0.1313
2026-09-02vercelCache read$0.13
2026-09-04vercelCache read$0.028
2026-09-11vercelCache read$0.13
2026-09-02azure_aiInput$0.58
2026-09-23azure_aiInput$0.58
2026-09-26azure_aiInput$0.58
2026-09-02edenaiInput$0.24
2026-09-04edenaiInput$0.32
2026-10-19edenaiInput$0.24
2026-10-20edenaiInput$0.32
2026-10-22edenaiInput$0.24
2026-10-26edenaiInput$0.32
2026-10-27edenaiInput$0.24
2026-10-30edenaiInput$0.32
2026-09-02jiekouInput$0.28
2026-10-20jiekouInput$0.27
2026-10-21jiekouInput$0.28
2026-10-22jiekouInput$0.27
2026-10-24jiekouInput$0.28
2026-10-25jiekouInput$0.27
2026-10-26jiekouInput$0.28
2026-10-27jiekouInput$0.27
2026-10-30jiekouInput$0.28
2026-09-02kiloInput$0.269
2026-09-24kiloInput$0.27
2026-09-29kiloInput$0.1344
2026-09-30kiloInput$0.27
2026-10-08kiloInput$0.27
2026-10-18kiloInput$0.27
2026-10-21kiloInput$0.28
2026-10-22kiloInput$0.27
2026-10-23kiloInput$0.28
2026-10-25kiloInput$0.27
2026-10-26kiloInput$0.28
2026-10-27kiloInput$0.27
2026-10-28kiloInput$0.28
2026-10-29kiloInput$0.27
2026-10-30kiloInput$0.28
2026-10-31kiloInput$0.27
2026-09-02llmgateway-providersInput$0.26
2026-09-04llmgateway-providersInput$0.269
2026-09-24llmgateway-providersInput$0.26
2026-10-20llmgateway-providersInput$0.269
2026-10-21llmgateway-providersInput$0.26
2026-10-24llmgateway-providersInput$0.269
2026-10-25llmgateway-providersInput$0.26
2026-10-28llmgateway-providersInput$0.269
2026-10-29llmgateway-providersInput$0.26
2026-09-02merge-gatewayInput$0.50
2026-09-04merge-gatewayInput$0.58
2026-09-24merge-gatewayInput$0.58
2026-10-01merge-gatewayInput$0.27
2026-10-17merge-gatewayInput$0.27
2026-10-21merge-gatewayInput$0.28
2026-10-22merge-gatewayInput$0.28
2026-10-23merge-gatewayInput$0.56
2026-10-24merge-gatewayInput$0.27
2026-10-25merge-gatewayInput$0.28
2026-10-27merge-gatewayInput$0.56
2026-10-28merge-gatewayInput$0.28
2026-10-29merge-gatewayInput$0.56
2026-10-30merge-gatewayInput$0.28
2026-10-31merge-gatewayInput$0.27
2026-11-01merge-gatewayInput$0.56
2026-09-02nano-gptInput$0.28
2026-09-24nano-gptInput$1.10
2026-10-13nano-gptInput$0.48
2026-10-14nano-gptInput$1.10
2026-10-15nano-gptInput$0.8957
2026-10-16nano-gptInput$1.10
2026-10-19nano-gptInput$0.66
2026-10-21nano-gptInput$0.2088
2026-10-24nano-gptInput$0.2088
2026-10-26nano-gptInput$0.2088
2026-10-27nano-gptInput$0.66
2026-10-28nano-gptInput$0.2088
2026-10-31nano-gptInput$0.66
2026-11-01nano-gptInput$0.2838
2026-09-02nano-gpt · thinkingInput$0.28
2026-10-16nano-gpt · thinkingInput$0.2088
2026-09-02novita-aiInput$0.27
2026-09-04novita-aiInput$0.269
2026-09-24novita-aiInput$0.27
2026-10-10novita-aiInput$0.89
2026-09-02openrouterInput$0.269
2026-09-06openrouterInput$0.269
2026-09-24openrouterInput$0.27
2026-09-28openrouterInput$0.27
2026-09-29openrouterInput$0.1344
2026-09-30openrouterInput$0.27
2026-10-08openrouterInput$0.27
2026-10-09openrouterInput$0.27
2026-10-10openrouterInput$0.27
2026-10-18openrouterInput$0.27
2026-10-21openrouterInput$0.28
2026-10-22openrouterInput$0.27
2026-10-23openrouterInput$0.28
2026-10-25openrouterInput$0.27
2026-10-26openrouterInput$0.28
2026-10-27openrouterInput$0.27
2026-10-28openrouterInput$0.28
2026-10-29openrouterInput$0.27
2026-10-30openrouterInput$0.28
2026-10-31openrouterInput$0.27
2026-09-02poeInput$0.27
2026-10-22poeInput$0.2626
2026-09-02together_aiInput$1.25
2026-09-23together_aiInput$1.25
2026-09-24together_aiInput$1.25
2026-09-02tokengoInput$0.19
2026-09-04tokengoInput$0.2174
2026-09-24tokengoInput$0.19
2026-10-21tokengoInput$0.2174
2026-10-23tokengoInput$0.19
2026-10-24tokengoInput$0.2174
2026-10-27tokengoInput$0.19
2026-10-28tokengoInput$0.2174
2026-10-29tokengoInput$0.19
2026-10-30tokengoInput$0.2174
2026-11-01tokengoInput$0.19
2026-09-02vercelInput$0.25
2026-09-04vercelInput$0.28
2026-09-11vercelInput$0.62
2026-09-24vercelInput$0.25
2026-10-21vercelInput$0.62
2026-10-23vercelInput$0.25
2026-10-24vercelInput$0.62
2026-10-27vercelInput$0.25
2026-10-28vercelInput$0.62
2026-10-29vercelInput$0.25
2026-10-30vercelInput$0.62
2026-11-01vercelInput$0.25
2026-09-02zenmuxInput$0.28
2026-09-24zenmuxInput$0.22
2026-10-21zenmuxInput$0.28
2026-10-22zenmuxInput$0.22
2026-10-23zenmuxInput$0.293
2026-10-25zenmuxInput$0.22
2026-10-26zenmuxInput$0.293
2026-10-27zenmuxInput$0.22
2026-10-28zenmuxInput$0.293
2026-10-29zenmuxInput$0.22
2026-10-30zenmuxInput$0.293
2026-10-31zenmuxInput$0.22
2026-09-02azure_aiOutput$1.68
2026-09-23azure_aiOutput$1.68
2026-09-26azure_aiOutput$1.68
2026-09-02edenaiOutput$0.90
2026-09-04edenaiOutput$0.89
2026-10-19edenaiOutput$0.90
2026-10-20edenaiOutput$0.89
2026-10-22edenaiOutput$0.90
2026-10-26edenaiOutput$0.89
2026-10-27edenaiOutput$0.90
2026-10-30edenaiOutput$0.89
2026-09-02jiekouOutput$1.14
2026-10-20jiekouOutput$1.00
2026-10-21jiekouOutput$1.14
2026-10-22jiekouOutput$1.00
2026-10-24jiekouOutput$1.14
2026-10-25jiekouOutput$1.00
2026-10-26jiekouOutput$1.14
2026-10-27jiekouOutput$1.00
2026-10-30jiekouOutput$1.14
2026-09-02kiloOutput$0.40
2026-09-24kiloOutput$0.41
2026-09-29kiloOutput$0.2016
2026-09-30kiloOutput$0.41
2026-10-11kiloOutput$0.41
2026-10-18kiloOutput$0.41
2026-10-21kiloOutput$0.42
2026-10-22kiloOutput$0.41
2026-10-23kiloOutput$0.42
2026-10-25kiloOutput$0.41
2026-10-26kiloOutput$0.42
2026-10-27kiloOutput$0.41
2026-10-28kiloOutput$0.42
2026-10-29kiloOutput$0.41
2026-10-30kiloOutput$0.42
2026-10-31kiloOutput$0.41
2026-09-02llmgateway-providersOutput$0.38
2026-09-04llmgateway-providersOutput$0.40
2026-09-24llmgateway-providersOutput$0.38
2026-10-20llmgateway-providersOutput$0.40
2026-10-21llmgateway-providersOutput$0.38
2026-10-24llmgateway-providersOutput$0.40
2026-10-25llmgateway-providersOutput$0.38
2026-10-28llmgateway-providersOutput$0.40
2026-10-29llmgateway-providersOutput$0.38
2026-09-02merge-gatewayOutput$1.50
2026-09-04merge-gatewayOutput$1.68
2026-09-24merge-gatewayOutput$1.68
2026-10-01merge-gatewayOutput$1.10
2026-10-17merge-gatewayOutput$1.10
2026-10-21merge-gatewayOutput$0.40
2026-10-22merge-gatewayOutput$0.40
2026-10-23merge-gatewayOutput$1.68
2026-10-24merge-gatewayOutput$1.10
2026-10-25merge-gatewayOutput$0.40
2026-10-27merge-gatewayOutput$1.68
2026-10-28merge-gatewayOutput$0.40
2026-10-29merge-gatewayOutput$1.68
2026-10-30merge-gatewayOutput$0.40
2026-10-31merge-gatewayOutput$1.10
2026-11-01merge-gatewayOutput$1.68
2026-09-02nano-gptOutput$0.42
2026-09-24nano-gptOutput$2.50
2026-10-13nano-gptOutput$1.98
2026-10-14nano-gptOutput$2.50
2026-10-15nano-gptOutput$0.99
2026-10-16nano-gptOutput$2.50
2026-10-19nano-gptOutput$1.98
2026-10-21nano-gptOutput$0.3096
2026-10-24nano-gptOutput$0.3096
2026-10-26nano-gptOutput$0.3096
2026-10-27nano-gptOutput$1.98
2026-10-28nano-gptOutput$0.3096
2026-10-31nano-gptOutput$1.98
2026-11-01nano-gptOutput$0.8514
2026-09-02nano-gpt · thinkingOutput$0.42
2026-10-16nano-gpt · thinkingOutput$0.3096
2026-09-02novita-aiOutput$1.12
2026-09-04novita-aiOutput$0.40
2026-09-24novita-aiOutput$1.12
2026-10-10novita-aiOutput$0.89
2026-09-02openrouterOutput$0.40
2026-09-06openrouterOutput$0.40
2026-09-24openrouterOutput$0.41
2026-09-28openrouterOutput$0.41
2026-09-29openrouterOutput$0.2016
2026-09-30openrouterOutput$0.41
2026-10-10openrouterOutput$0.41
2026-10-11openrouterOutput$0.41
2026-10-18openrouterOutput$0.41
2026-10-21openrouterOutput$0.42
2026-10-22openrouterOutput$0.41
2026-10-23openrouterOutput$0.42
2026-10-25openrouterOutput$0.41
2026-10-26openrouterOutput$0.42
2026-10-27openrouterOutput$0.41
2026-10-28openrouterOutput$0.42
2026-10-29openrouterOutput$0.41
2026-10-30openrouterOutput$0.42
2026-10-31openrouterOutput$0.41
2026-09-02poeOutput$0.40
2026-10-22poeOutput$0.3838
2026-09-02together_aiOutput$1.25
2026-09-23together_aiOutput$1.25
2026-09-24together_aiOutput$1.25
2026-09-02tokengoOutput$0.71
2026-09-04tokengoOutput$0.326
2026-09-24tokengoOutput$0.71
2026-10-21tokengoOutput$0.326
2026-10-23tokengoOutput$0.71
2026-10-24tokengoOutput$0.326
2026-10-27tokengoOutput$0.71
2026-10-28tokengoOutput$0.326
2026-10-29tokengoOutput$0.71
2026-10-30tokengoOutput$0.326
2026-11-01tokengoOutput$0.71
2026-09-02vercelOutput$0.95
2026-09-04vercelOutput$0.42
2026-09-11vercelOutput$1.85
2026-09-24vercelOutput$0.95
2026-10-21vercelOutput$1.85
2026-10-23vercelOutput$0.95
2026-10-24vercelOutput$1.85
2026-10-27vercelOutput$0.95
2026-10-28vercelOutput$1.85
2026-10-29vercelOutput$0.95
2026-10-30vercelOutput$1.85
2026-11-01vercelOutput$0.95
2026-09-02zenmuxOutput$0.43
2026-09-24zenmuxOutput$0.33
2026-10-21zenmuxOutput$0.43
2026-10-22zenmuxOutput$0.33
2026-10-23zenmuxOutput$0.4395
2026-10-25zenmuxOutput$0.33
2026-10-26zenmuxOutput$0.4395
2026-10-27zenmuxOutput$0.33
2026-10-28zenmuxOutput$0.4395
2026-10-29zenmuxOutput$0.33
2026-10-30zenmuxOutput$0.4395
2026-10-31zenmuxOutput$0.33

More from DeepSeek

most-hosted first