Grok 4.20 (non-reasoning)

Grok 4.20 (non-reasoning) is the non-reasoning snapshot of xAI's Grok 4.20, built for fast, precise responses with agentic tool calling and strict prompt adherence.

start building →

compare models

                                                      ..     .
                                                  .:-+-++++--+---:.
                                                 :-=%@@@@@@@%@@%%+.
                                               .:+%@@@@@@@@@@@@%-:
                                              .:+@@@@@@@@@@@@%=-..
                                             .:*@@@@@@@@@@@@@=-:.+-.
                                           ..:=@@@@@@@@@@%%%*-:-==-.
                                           .-=@@@@@@@@@@@%%*::-=%%-..
                                          .+%@@@@@@@@@@@@%+::-=%@%-.
                                        .:+%@@@@@@@@@@@%%+::+%%%@%+.
                                        :*%@@@@@@@@@%@%=-::*%%%@%%-.
                                      .:=@@@@@@@@@@@@@*:::=%%@%%%%-..
                                    ..-=@@@@@@@@@@@@%*:::*%%@@%%%%-.
                                    .-%@@@@@@@@@@@%%*::-=%%@%@%%%%+.
                                   :+%@@@@@@@@@@@%%+:.-=%@@@@@@%%%-.
      .... .  .                   :+%@@@@@@@@@@%%%-..:*%%%%%@%%%%=-.
 .--++++++--+++--..             .:*%@@@@@@@@@@@%=:. ..*%%@%%@%%%%%-.
 .+@@@@@@@@@@@@@%-:.           .:=@@@@@@@%%@@%%*:..  :*%@%%%%%%%%=-.
  :+%@##@##@@@@@@%+:          .:=%@@@@@@@@@@%%*.     :*%%%%%%%%%%=-.
   :+%@@##@##@@@@@%*:.       .-%%@@@@@@@@@@%%+..     :*%%%%%%%%%%=-.
    .-=@@@##@@@@@@@@=:.     .+%%@@@@@@%@%@%%-..     .:*%%%%%%%%%%=-.
     .:=@@#@@#@@@@@@%=-.. .:*%@@@@@@@@@%%%=-.        .+=%%%%%%%%==-.
      .:*%@#@@@@@@@@@@%-..:*%%@@@@@@%%%%%=:..        .*%%%%%%%%%%=-.
        :+@@@@@@@@@@@@@%+::+%@@@@%@@%%%%*:.          :*=%%%%%%%%%=-.
        .:-%@@@@@@@@@@@@%+::+%@@@@@%%%%*:.           :+=%%%%%%====-.
          .-%@@@@@@@@@@@%%*::-=%%%%@%=+:             .*=%%%=%%%==*-.
           .:=%@@@@@@@@@@@%=:::*%%%%%-.             ..+=%%%%%=%===-.
           . :*%@@@@@@@@%@%%=:::*%%=-.               .+=%%=%%%===*-.
              :+%@@@@@@%@@@%@%-.:+*:..              ..*=%=%===%===-.
              ..-%@@@@@@@@%%@%%-:..                  .+=======%==*:.
                .-%%@@%@%%@%%@%=+:                   :+==%%=%=====-.
                .:-=%%@%%@@@%%%%%*:..                .+====%=====*-.
             ..+*-::*%%@%%@%@@@%%%*:.               ..+===========:.
            .:*%%=-::*%%%@%%%%%%%%%*-.              ..+==========*-.
           .:=%@@%=-::+%%%%%%%@@%%%%=-.             ..+==========*-.
          .-=%@@@@%%+:.-=%%%%%%%%%%%%=+.             .+==========*:..
        ..-=%@%@@@@%%+::-*%%%%%%%%%%%==+:           ..+*=======*=*:.
        .+%%%@%@%%@%@%*:.:*=%%%%%%%%%===+:          ..+=========**-.
       :+%%%@%%%@%@%%==:..:+=%%%%%%%%%%%=*:.         .+=========**-
     .:*%%%%@@%%%@%%%*:.  ..+=%%%%%%%=%%%=*:.        .+**==*===***:
    .:=%%%@@%@%%%@%%*:.     .-==%%%%%%%%=%=*-.       .+=*=*==*=***:.
   .-=%@@%%@@@@%%%=*..       .:==%=%%%%=%=%==-.      .-==***=*****:.
  .-%%@@@@%@%%%%%=-.          .:*=%%==%======*-.    ..+**=*******+:.
..+=%%%%%%%%%%%==-.            .:+========*****+..   .-*********++:.
 .:-------------:.              ..::---::-:-:-::.    ..:::::::::::.
  .  . . .                        .  .    .                    .

input · usd / 1m tok

$1.25

output · usd / 1m tok

$2.50

context window

1mtokens

modalities · in → out

pricing

base api ratesusd per 1m tokens

input: $1.25over 200k · $2.50
output: $2.50over 200k · $5.00
cache read: $0.20over 200k · $0.40

brackets are by input tokens per request: a request over a threshold is billed at that bracket's rates for all of its tokens. cache creation can have a separate charge.

providers

xAI hosts Grok 4.20 (non-reasoning). Impossibl routes your request intelligently and handles failover automatically when a provider is down.

specifications

takes text and image in, gives text outxai/grok-4.20-0309-non-reasoning

creatorxAI
model typeText generation
context window1,000,000 tokens
maximum outputnot published
input
takes text and image in, gives text out
outputtext
tool callingnot verified
releasedmar 10, 2026
endpoint/v1/chat/completions

unknown means the public catalog does not establish that fact. a model's documented capabilities can differ from what a particular api supports. see the model documentation for integration details.

try it · your prompt, this model, live

# responses API
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["IMPOSSIBL_API_KEY"],
    base_url="https://api.impossibl.com/v1",
)

resp = client.responses.create(
    model="xai/grok-4.20-0309-non-reasoning",
    input="▌",
)
print(resp.output_text)

edit the prompt, then run

# responses API
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["IMPOSSIBL_API_KEY"],
    base_url="https://api.impossibl.com/v1",
)

resp = client.responses.create(
    model="xai/grok-4.20-0309-non-reasoning",
    input="▌",
)
print(resp.output_text)

edit the prompt, then run

benchmarks

intelligence index

14.6#88 of 108 models on the gateway

omniscience-42.6#77 of 84
gpqa78.5%#58 of 80
humanity’s last exam24.5%#55 of 87

GPQA78.5%
Tau269.6%
MMMU-Pro64.0%
IFBench47.8%
AA-LCR25.3%
Humanity’s Last Exam24.5%
Terminal-Bench Hard22.0%
CritPt0.0%

Intelligence Index14.61
Omniscience Index-42.58
intelligence index (API)14.6

view source ↗all source data and breakdowns (json) ↗

each configuration is listed separately; these evaluations describe the tested model configuration, and performance and costs can differ on impossibl. unmeasured scores are omitted and api values keep their source units. last checked 2026-10-14.

compare Grok 4.20 (non-reasoning) with other models

frontier labs and open weights, best-ranked firstusd / 1m tok · in / out

anthropic/claude-opus-5-5$4.00 / $20.00compare →
openai/gpt-6-astra$10.00 / $50.00compare →
meta/muse-wonder$4.00 / $20.00compare →
xiaomi/mimo-v2.6-pro$0.435 / $0.87compare →

impossibl models --all →

frequently asked questions

What is Grok 4.20 (non-reasoning)?

Grok 4.20 (non-reasoning) is the non-reasoning snapshot of xAI's Grok 4.20, built for fast, precise responses with agentic tool calling and strict prompt adherence. It shares the reasoning variant's text and image inputs, function calling, and structured outputs, but skips extended reasoning for lower latency.

How much does Grok 4.20 (non-reasoning) cost?

Grok 4.20 (non-reasoning) costs $1.25 per million input tokens and $2.50 per million output tokens through the Impossibl API. Cache reads cost $0.20 per million tokens. These base rates apply through 200,000 total input tokens per request. Above 200,000 input tokens, the long-context rates in the pricing table apply to the whole request. All rates are in USD.

view the pricing table ↑

What is the context length of Grok 4.20 (non-reasoning)?

Grok 4.20 (non-reasoning) has a 1,000,000-token context window through the Impossibl API.

What inputs and outputs does Grok 4.20 (non-reasoning) support?

Grok 4.20 (non-reasoning) accepts text and image as input. Grok 4.20 (non-reasoning) returns text.

How do I use Grok 4.20 (non-reasoning) through an API?

Grok 4.20 (non-reasoning) is available through the Impossibl API. Use xai/grok-4.20-0309-non-reasoning as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request.

follow the api quickstart →

When was Grok 4.20 (non-reasoning) released?

xAI released Grok 4.20 (non-reasoning) on March 10, 2026.

Creator release announcement ↗

this page describes Grok 4.20 (non-reasoning) as listed by the impossibl public api. prices and availability come from that catalog, refreshed here every five minutes. this is an impossibl offering, not a comparison of independent hosting-provider quotes. model descriptions and release dates, where available, use the creator sources linked under the answers above.

view the source catalog ↗model as markdown ↗report a correction →