DeepSeek: DeepSeek V3 0324

DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) mo…

deepseek/deepseek-chat-v3-0324

DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...

Specifications & capabilities

Input
text
Output
text
Context window · tokens
163840
Knowledge cutoff
2024-07-31
Published model ID
DeepSeek-V3-0324
Capabilities
promptCaching, structuredOutputs, toolCalling

Evaluations

21–40 / 84

OpenRouter Benchmarks

53.325%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DeepInfra
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
64
sample size
391
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T20:34:10.117Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
391
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.5332481099744245
accuracy_std_dev
0.020375000000000032
cost_per_task_usd
0.004206966751918158
output_tokens_per_task
4524.092071611253
latency_seconds_per_task
160.59066751918158
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
71
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-chat-v3-0324
source model id
deepseek/deepseek-chat-v3-0324
Original result ↗

OpenRouter Benchmarks

45%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · auto-routing
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
112
sample size
100
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T16:11:46.632Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
100
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.45
accuracy_std_dev
0.0033330000000000026
cost_per_task_usd
0.07979226
output_tokens_per_task
21258.48
latency_seconds_per_task
840.71505
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
128
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-chat-v3-0324
source model id
deepseek/deepseek-chat-v3-0324
Original result ↗

OpenRouter Benchmarks

54.63%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · auto-routing
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
120
sample size
396
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
2
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-08-29T20:57:58.443Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
396
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.5462965
accuracy_std_dev
0.08670050000000001
cost_per_task_usd
0.006357126262626262
output_tokens_per_task
5650.492424242424
latency_seconds_per_task
124.53147474747475
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
132
benchmark_evaluation
true
openrouter_permaslug
deepseek/deepseek-chat-v3-0324
source model id
deepseek/deepseek-chat-v3-0324
Original result ↗

LMArena

1,389.87

source id
lmarena
leaderboard name
LMArena Text Style Control Creative Writing
dimension
style
metric
Arena rating
rank
130
sample size
5982
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1381.769859813475
confidence upper
1397.9697693102469
Original result ↗

LMArena

1,383.153

source id
lmarena
leaderboard name
LMArena Text Style Control Industry Entertainment And Sports And Media
dimension
style
metric
Arena rating
rank
141
sample size
8000
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1376.156888722469
confidence upper
1390.1499076750406
Original result ↗

LMArena

1,333.258

source id
lmarena
leaderboard name
LMArena Text Japanese
dimension
general
metric
Arena rating
rank
144
sample size
1056
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1314.3181831455295
confidence upper
1352.1973576924597
Original result ↗

LMArena

1,396.117

source id
lmarena
leaderboard name
LMArena Text Style Control Polish
dimension
style
metric
Arena rating
rank
145
sample size
2735
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1383.999686635223
confidence upper
1408.2337059449292
Original result ↗

LMArena

1,391.601

source id
lmarena
leaderboard name
LMArena Text Style Control German
dimension
style
metric
Arena rating
rank
145
sample size
1224
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1373.94913985116
confidence upper
1409.2527873864192
Original result ↗

LMArena

1,343.618

source id
lmarena
leaderboard name
LMArena Text Style Control Japanese
dimension
style
metric
Arena rating
rank
145
sample size
1056
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1324.512690110648
confidence upper
1362.723573006812
Original result ↗

LMArena

1,364.215

source id
lmarena
leaderboard name
LMArena Text Creative Writing
dimension
general
metric
Arena rating
rank
153
sample size
5982
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1356.234580588784
confidence upper
1372.1948825609377
Original result ↗

LMArena

1,410.261

source id
lmarena
leaderboard name
LMArena Text Style Control Multi Turn
dimension
style
metric
Arena rating
rank
154
sample size
7673
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1403.015866977495
confidence upper
1417.5060512139307
Original result ↗

LMArena

1,386.314

source id
lmarena
leaderboard name
LMArena Text Style Control Industry Writing And Literature And Language
dimension
style
metric
Arena rating
rank
154
sample size
10270
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1379.8568608800958
confidence upper
1392.7703415506126
Original result ↗

LMArena

1,400.55

source id
lmarena
leaderboard name
LMArena Text Style Control Russian
dimension
style
metric
Arena rating
rank
155
sample size
3570
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1390.9616272414137
confidence upper
1410.1378276900502
Original result ↗

LMArena

1,373.714

source id
lmarena
leaderboard name
LMArena Text German
dimension
general
metric
Arena rating
rank
156
sample size
1224
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1356.2668026963233
confidence upper
1391.1619395371354
Original result ↗

LMArena

1,427.023

source id
lmarena
leaderboard name
LMArena Text Style Control Industry Medicine And Healthcare
dimension
style
metric
Arena rating
rank
157
sample size
2657
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1415.2829960837846
confidence upper
1438.7635535833335
Original result ↗

LMArena

1,415.274

source id
lmarena
leaderboard name
LMArena Text Style Control French
dimension
style
metric
Arena rating
rank
157
sample size
593
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1390.5385692920295
confidence upper
1440.0088208049322
Original result ↗

LMArena

1,339.061

source id
lmarena
leaderboard name
LMArena Text Style Control Korean
dimension
style
metric
Arena rating
rank
158
sample size
964
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1319.353401408107
confidence upper
1358.7678125669559
Original result ↗

LMArena

1,409.551

source id
lmarena
leaderboard name
LMArena Text Style Control Industry Legal And Government
dimension
style
metric
Arena rating
rank
159
sample size
2906
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1398.3761804362543
confidence upper
1420.7256063191896
Original result ↗

LMArena

1,368.636

source id
lmarena
leaderboard name
LMArena Text Polish
dimension
general
metric
Arena rating
rank
161
sample size
2735
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1356.6394659255989
confidence upper
1380.6333574581276
Original result ↗

LMArena

1,320.438

source id
lmarena
leaderboard name
LMArena Text Korean
dimension
general
metric
Arena rating
rank
161
sample size
964
evaluated at
2026-11-09T00:00:00.000Z
source model id
DeepSeek-V3 (0324)
confidence lower
1300.8303457342104
confidence upper
1340.0458259153397
Original result ↗

Published prices

1–3 / 3

Provider specifications & access

1–2 / 2

deepseek-ai/DeepSeek-V3-0324 ↗

source model id
publishers/deepseek-ai/models/deepseek-v3@deepseek-v3-0324
hosting
vertex-self-deployed
platforms
  • Google Cloud Vertex AI
architecture
input_modalities
  • text
output_modalities
  • text
capabilities

    deepseek-ai/DeepSeek-V3-0324 ↗

    source model id
    publishers/hf-deepseek-ai/models/deepseek-v3-0324@001
    hosting
    vertex-self-deployed
    platforms
    • Google Cloud Vertex AI
    architecture
    input_modalities
    • text
    output_modalities
    • text
    capabilities

      Safety assessments

      1–1 / 1

      Foundation Model Transparency Index 2025 ↗

      32 / 100

      source id
      stanfordFMTI
      canonical model id
      deepseek/deepseek-chat-v3-0324
      taxonomy
      transparency
      hazard category
      transparency
      hazard name
      Organization transparency
      policy area
      DeepSeek
      assessment type
      organization-level disclosure
      metric
      disclosure_score
      result label
      32 / 100
      sample size
      100
      updated at
      2026-11-09T18:08:05.227Z

      Source specification observations

      1–3 / 3

      openRouter ↗

      field
      context_length
      value
      163840
      normalized value
      163840
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      knowledge_cutoff
      reported field
      knowledge_cutoff
      value
      2024-07-31
      normalized value
      2024-07-31
      precision
      day
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      max_output_tokens
      value
      115200
      normalized value
      115200
      scope
      source-model
      source id
      openRouter
      confidence
      1

      Original sources

      1–15 / 15

      Hugging Face · deepseek-ai/DeepSeek-V3-0324 ↗

      Google Cloud · Model page ↗

      Google Cloud · Model page ↗

      OpenRouter · Model page ↗

      Google Cloud · Model documentation ↗

      arXiv · Paper · 2412.19437 ↗

      GitHub · Catalog ↗

      GitHub · Catalog ↗

      Google Cloud · Catalog ↗

      Hugging Face · Catalog ↗

      Hugging Face · Catalog ↗

      LMArena · Catalog ↗

      OpenRouter · Catalog ↗

      OpenRouter · Catalog ↗

      livebench.ai · Catalog ↗

      Source links

      1–10 / 10

      Hugging Face model card

      1–1 / 1

      Explore more models

      Models

      Loading
      EVIDENCE / COMPARISON

      Benchmarks

      Explore every benchmark

      Loading

      These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

      Install TensorTrace