gpt·basedElo-ranked, value-sorted
Sign in

Stay updated

Get an email when a cheaper, stronger model than yours appears on the leaderboard. No spam.

gpt·based

The LLM leaderboard that tells you when to switch. Arena rankings joined with OpenRouter pricing, refreshed daily.

Use cases

  • Best for coding
  • Best for multilingual
  • Cheapest worth using
  • Compare two models

Browse

  • Text models
  • WebDev models
  • Vision models
  • Image generation
  • Image edit

Data sources

  • LMArena
  • Providers

Developers

  • API on RapidAPI
  • OpenAPI spec

Contact

  • Contact page
  • [email protected]
© 2026 gptbasedDaily refresh, never paid placement.
    gpt·basedElo-ranked, value-sorted
    Sign in
    Back to Text · Coding

    DeepSeek: DeepSeek V4 Flash

    Released Apr 24, 2026
    text→text
    Set alert
    Ranked in:Text
    Try it on:
    OOpenRouter

    DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and...

    Best overall rank
    #70
    ▼ 2 vs 7d ago
    Total arena votes
    599,295
    Across 58 (subset, category) entries
    Providers on OR
    18
    DeepInfra + others
    First seen
    2026-05-08
    72 days tracked

    Trend

    Pricing

    Per token

    Input (per 1M)
    $0.098
    Output (per 1M)
    $0.196
    Cache Read (per 1M)
    $0.020

    Capabilities

    Modality
    text->text
    Context
    1M tokens
    Tokenizer
    DeepSeek
    Moderation
    Unmoderated
    Hugging Face
    deepseek-ai/DeepSeek-V4-Flash
    Supported parameters
    frequency_penalty
    include_reasoning
    logit_bias
    logprobs
    max_tokens
    min_p
    presence_penalty
    reasoning
    reasoning_effort
    repetition_penalty
    response_format
    seed
    stop
    structured_outputs
    temperature
    tool_choice
    tools
    top_k
    top_logprobs
    top_p

    Providers

    ProviderRegion / TagContextInput/MOutput/MUptime 1dLatencyQuantCache
    DeepInfradeepinfra/fp41M$0.090$0.18099.94%-
    fp4
    -
    Baidubaidu/fp81M$0.098$0.19799.94%-
    fp8
    -
    DeepSeekdeepseek1M$0.140$0.28099.94%--✓
    Alibabaalibaba1M$0.134$0.26899.82%---
    Io Netio-net/fp81M$0.250$0.39999.79%-
    fp8
    -
    DigitalOceandigitalocean262K$0.112$0.22499.74%---
    SiliconFlowsiliconflow/fp81M$0.130$0.28099.72%-
    fp8
    -
    WandBwandb/fp81M$0.140$0.28099.61%-
    fp8
    -
    AtlasCloudatlas-cloud/fp41M$0.140$0.28099.55%-
    fp4
    -
    AkashMLakashml/fp8131K$0.140$0.28099.36%-
    fp8
    -
    Parasailparasail/fp81M$0.140$0.28099.34%-
    fp8
    -
    Venicevenice1M$0.138$0.27599.30%---
    Novitanovita/fp81M$0.140$0.28099.15%-
    fp8
    -
    Ambientambient/fp41M$0.500$0.80098.87%-
    fp4
    -
    GMICloudgmicloud/fp81M$0.098$0.19698.78%-
    fp8
    -
    StreamLakestreamlake/fp81M$0.097$0.19398.78%-
    fp8
    -
    Morphmorph1M$0.139$0.27898.39%---
    Fireworksfireworks1M$0.140$0.28097.37%---

    Per-category ranks

    Text

    RankCategoryVariantTrendEloVotes
    #39Japanese
    thinking
    1425478
    #51Industry Legal And Government
    thinking
    14623,364
    #53Korean
    -
    1402698
    #55Chinese
    thinking
    14882,090
    #58Multi Turn
    -
    14547,740
    #59German
    thinking
    1441717
    #60Expert
    thinking
    14714,100
    #60Instruction Following
    thinking
    143514,453
    #60Longer Query
    -
    145118,583
    #61French
    thinking
    14631,640
    #61Industry Life And Physical And Social Science
    thinking
    14626,997
    #63Math
    thinking
    14392,135
    #63Spanish
    -
    14371,292
    #64Creative Writing
    -
    14116,874
    #64Industry Life And Physical And Social Science
    -
    14606,796
    #64Japanese
    -
    1393508
    #64Longer Query
    thinking
    145018,641
    #64Non English
    -
    142123,270
    #64Polish
    -
    1435861
    #66Industry Entertainment And Sports And Media
    -
    14089,023
    #66Industry Medicine And Healthcare
    thinking
    14633,058
    #66Industry Writing And Literature And Language
    -
    142110,198
    #66Russian
    thinking
    14404,679
    #67English
    thinking
    144919,293
    #67Hard Prompts
    -
    146028,308
    #67Industry Legal And Government
    -
    14513,397
    #67Instruction Following
    -
    142914,391
    #68Polish
    thinking
    1429889
    #69Exclude Ties
    thinking
    143532,458
    #69Industry Medicine And Healthcare
    -
    14612,992
    #69Industry Writing And Literature And Language
    thinking
    141910,348
    #69Multi Turn
    thinking
    14477,600
    #70Industry Mathematical
    thinking
    14412,220
    #70Overall
    thinking
    143842,708
    #71Creative Writing
    thinking
    14076,981
    #71Hard Prompts English
    -
    146313,632
    #71Industry Software And It Services
    thinking
    147417,466
    #71Russian
    -
    14374,666
    #72Coding
    -
    148212,454
    #72Expert
    -
    14604,106
    #72Hard Prompts
    thinking
    145928,280
    #72Industry Entertainment And Sports And Media
    thinking
    14059,227
    #72Overall
    -
    143842,807
    #72Spanish
    thinking
    14271,231
    #73English
    -
    144619,486
    #73Exclude Ties
    -
    143332,372
    #73Hard Prompts English
    thinking
    146313,430
    #75Chinese
    -
    14752,087
    #75German
    -
    1423690
    #75Industry Business And Management And Financial Operations
    -
    14378,583
    #76Coding
    thinking
    148212,406
    #76Industry Software And It Services
    -
    147117,477
    #79Industry Business And Management And Financial Operations
    thinking
    14338,782
    #79Non English
    thinking
    141023,370
    #80Korean
    thinking
    1375765
    #82French
    -
    14451,665
    #83Industry Mathematical
    -
    14362,348
    #85Math
    -
    14272,185