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 · Polish

    OpenAI: gpt-oss-20b (free)

    Released Aug 5, 2025
    openai/gpt-oss-20b:free
    text→text
    Set alert
    Ranked in:Text
    Try it on:
    OOpenRouter

    gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...

    Best overall rank
    #222
    ▼ 1 vs 7d ago
    Total arena votes
    63,117
    Across 28 (subset, category) entries
    Providers on OR
    2
    OpenInference + others
    First seen
    2026-05-08
    61 days tracked

    Trend

    Pricing

    No priced fields.

    Capabilities

    Modality
    text->text
    Context
    131K tokens
    Tokenizer
    GPT
    Knowledge cutoff
    2024-06-30
    Max output
    33K tokens
    Moderation
    Unmoderated
    Hugging Face
    openai/gpt-oss-20b
    Supported parameters
    frequency_penalty
    include_reasoning
    logprobs
    max_tokens
    min_p
    presence_penalty
    reasoning
    repetition_penalty
    response_format
    seed
    stop
    structured_outputs
    temperature
    tool_choice
    tools
    top_a
    top_k
    top_logprobs
    top_p

    Providers

    ProviderRegion / TagContextInput/MOutput/MUptime 1dLatencyQuantCache
    OpenInferenceopen-inference/int8131K--99.27%-
    int8
    -
    Darkbloomdarkbloom131K--96.68%---

    Per-category ranks

    Text

    RankCategoryVariantTrendEloVotes
    #145Korean
    -
    1293218
    #151Japanese
    -
    1263233
    #183Math
    -
    1336678
    #185Industry Mathematical
    -
    1330583
    #185Industry Medicine And Healthcare
    -
    1353561
    #187Polish
    -
    1277780
    #189Spanish
    -
    1304253
    #190Chinese
    -
    1349488
    #196Expert
    -
    1318490
    #198Coding
    -
    13702,168
    #199Industry Business And Management And Financial Operations
    -
    13221,847
    #201Industry Software And It Services
    -
    13583,708
    #203German
    -
    1270208
    #215Non English
    -
    12995,548
    #215Russian
    -
    1301605
    #217Hard Prompts English
    -
    13392,448
    #219Exclude Ties
    -
    12627,780
    #222Hard Prompts
    -
    13234,797
    #222Overall
    -
    131710,625
    #223English
    -
    13315,065
    #223Industry Life And Physical And Social Science
    -
    13291,686
    #234Industry Legal And Government
    -
    1322694
    #235Multi Turn
    -
    12921,838
    #237Longer Query
    -
    13012,091
    #239Instruction Following
    -
    12812,555
    #245Industry Entertainment And Sports And Media
    -
    12571,746
    #245Industry Writing And Literature And Language
    -
    12742,181
    #259Creative Writing
    -
    12381,243

    Related models

    OpenAI: gpt-oss-20bopenai/gpt-oss-20b