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 · English

    Z.ai: GLM 5.2

    Released Jun 16, 2026
    text→text
    Set alert
    Ranked in:TextWebDev
    Try it on:
    OOpenRouter

    GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...

    Best overall rank
    #4
    ▼ 1 vs 7d ago
    Total arena votes
    123,517
    Across 33 (subset, category) entries
    Providers on OR
    30
    Wafer + others
    First seen
    2026-06-17
    32 days tracked

    Trend

    Pricing

    Per token

    Input (per 1M)
    $0.253
    Output (per 1M)
    $0.796
    Cache Read (per 1M)
    $0.047

    Capabilities

    Modality
    text->text
    Context
    1M tokens
    Tokenizer
    Other
    Max output
    131K tokens
    Moderation
    Unmoderated
    Hugging Face
    zai-org/GLM-5.2
    Supported parameters
    frequency_penalty
    include_reasoning
    logit_bias
    logprobs
    max_tokens
    min_p
    parallel_tool_calls
    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
    Waferwafer/fast1M$3.00$10.2599.99%-
    fp4
    -
    SiliconFlowsiliconflow/fp81M$1.30$4.0999.95%-
    fp8
    -
    Waferwafer/fp41M$1.20$4.1099.90%-
    fp4
    -
    WandBwandb/fp4262K$1.39$4.4099.84%-
    fp4
    -
    Alibabaalibaba1M$1.24$3.8999.79%---
    AtlasCloudatlas-cloud/fp81M$1.26$3.9699.78%-
    fp8
    -
    Fireworksfireworks/fast1M$2.10$6.6099.77%---
    Friendlifriendli1M$1.40$4.4099.77%---
    Decartdecart/fp41M$1.20$4.2099.74%-
    fp4
    -
    StreamLakestreamlake/fp81M$0.254$0.79999.73%-
    fp8
    -
    Io Netio-net/fp8262K$2.34$5.1299.70%-
    fp8
    -
    Togethertogether262K$1.40$4.4099.70%---
    Novitanovita/fp81M$0.253$0.79699.68%-
    fp8
    -
    Cloudflarecloudflare262K$1.40$4.4099.68%---
    Parasailparasail/fp4262K$1.40$4.4099.61%-
    fp4
    -
    DeepInfradeepinfra/fp41M$0.930$3.0099.42%-
    fp4
    -
    GMICloudgmicloud/fp81M$1.40$4.4099.41%-
    fp8
    -
    Baidubaidu/fp81M$0.280$0.88099.39%-
    fp8
    -
    Ambientambient/fp8101K$1.05$4.4099.34%-
    fp8
    -
    BaseTenbaseten/fp8256K$1.40$4.4098.93%-
    fp8
    -
    Venicevenice/fp81M$1.40$4.4098.90%-
    fp8
    -
    DigitalOceandigitalocean262K$1.05$4.4098.61%---
    Morphmorph1M$1.10$4.1098.54%---
    AkashMLakashml/fp897K$1.30$4.4098.13%-
    fp8
    -
    Fireworksfireworks1M$1.40$4.4098.11%---
    Inceptroninceptron/fp41M$0.940$2.9098.07%-
    fp4
    -
    Chuteschutes/fp41M$1.40$4.4097.82%-
    fp4
    -
    Phalaphala1M$1.40$4.4096.61%---
    Ionstreamionstream/fp41M$1.40$4.4095.88%-
    fp4
    -
    Z.AIz-ai/fp81M$1.40$4.4093.95%-
    fp8
    -

    Per-category ranks

    Text

    RankCategoryVariantTrendEloVotes
    #6Spanish
    harness: max
    1488490
    #11German
    harness: max
    1482271
    #11Japanese
    harness: max
    1476201
    #13Industry Mathematical
    harness: max
    1496849
    #19Math
    harness: max
    1478770
    #21Industry Life And Physical And Social Science
    harness: max
    14902,619
    #23Non English
    harness: max
    14528,364
    #24Instruction Following
    harness: max
    14645,373
    #25Chinese
    harness: max
    1512847
    #26English
    harness: max
    14777,263
    #26Hard Prompts English
    harness: max
    14924,978
    #27Korean
    harness: max
    1429265
    #28French
    harness: max
    1485596
    #28Industry Medicine And Healthcare
    harness: max
    14871,181
    #29Creative Writing
    harness: max
    14472,804
    #29Industry Software And It Services
    harness: max
    15036,168
    #30Exclude Ties
    harness: max
    147511,572
    #30Longer Query
    harness: max
    14797,039
    #30Overall
    harness: max
    146815,642
    #31Hard Prompts
    harness: max
    149010,288
    #32Industry Legal And Government
    harness: max
    14761,282
    #32Industry Writing And Literature And Language
    harness: max
    14523,947
    #32Russian
    harness: max
    14681,538
    #33Coding
    harness: max
    15124,382
    #34Industry Entertainment And Sports And Media
    harness: max
    14383,721
    #35Expert
    harness: max
    14891,625
    #41Multi Turn
    harness: max
    14712,578
    #47Industry Business And Management And Financial Operations
    harness: max
    14553,172
    #66Polish
    harness: max
    1429275

    WebDev

    RankCategoryVariantTrendEloVotes
    ★ #4Overall
    harness: max
    15874,722
    ★ #4Webdev
    harness: max
    15874,722
    ★ #4Webdev-react
    harness: max
    15973,436
    #11Webdev-html
    harness: max
    1531537