Qwen3.8 2.4T A95B

Qwen3.8 2.4T A95B is the open-weight flagship of Alibaba's Qwen3.8 family, a sparse mixture-of-experts vision-language model and the open release of Qwen3.8 Max.

start building →

compare models

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

input · usd / 1m tok

$1.20

output · usd / 1m tok

$1.20

context window

262.14ktokens

modalities · in → out

pricing

base api ratesusd per 1m tokens

input: $1.20
output: $1.20
cache read: $0.60

providers

Fireworks hosts Qwen3.8 2.4T A95B. Impossibl routes your request intelligently and handles failover automatically when a provider is down.

Fireworks

specifications

takes text in, gives text outqwen/qwen3.8-2.4t-a95b

creatorQwen
model typeText generation
context window262,144 tokens
maximum outputnot published
input
takes text in, gives text out
outputtext
tool callingnot verified
releasedaug 12, 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="qwen/qwen3.8-2.4t-a95b",
    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="qwen/qwen3.8-2.4t-a95b",
    input="▌",
)
print(resp.output_text)

edit the prompt, then run

benchmarks

intelligence index

40.0#16 of 90 models on the gateway

coding index71.9#17 of 60
agentic index50.4#11 of 49
omniscience4.3#29 of 86
gpqa93.5%#8 of 88
humanity’s last exam42.4%#20 of 89
terminal-bench v2.182.0%#17 of 57

GPQA93.5%
Terminal-Bench v2.182.0%
AA-LCR80.3%
AutomationBench Partial57.2%
SciCode54.1%
Tau Banking49.1%
Humanity’s Last Exam42.4%
CritPt20.0%
GDP.pdf All-pass15.6%
Terminal-Bench v4.011.1%
Multilingual LCR0.0%

Intelligence Index40.04
Omniscience Index4.32
GDPval Elo1627.82
coding index (API)71.9
agentic index (API)50.4
intelligence index (API)40

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 2026-09-11.

compare Qwen3.8 2.4T A95B with other models

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

anthropic/claude-fable-5-1$10.00 / $50.00compare →
openai/gpt-6-astra$10.00 / $50.00compare →
meta/muse-spark-1.3$1.25 / $4.25compare →
zai/glm-5.3$1.40 / $4.40compare →

impossibl models --all →

frequently asked questions

What is Qwen3.8 2.4T A95B?

Qwen3.8 2.4T A95B is the open-weight flagship of Alibaba's Qwen3.8 family, a sparse mixture-of-experts vision-language model and the open release of Qwen3.8 Max. It targets coding, productivity, research, and long-horizon agent tasks, with stronger autonomous planning and better handling of environment feedback.

How much does Qwen3.8 2.4T A95B cost?

Qwen3.8 2.4T A95B costs $1.20 per million input tokens and $1.20 per million output tokens through the Impossibl API. Cache reads cost $0.60 per million tokens. All rates are in USD.

view the pricing table ↑

What is the context length of Qwen3.8 2.4T A95B?

Qwen3.8 2.4T A95B has a 262,144-token context window through the Impossibl API.

What inputs and outputs does Qwen3.8 2.4T A95B support?

Qwen3.8 2.4T A95B accepts text as input. Qwen3.8 2.4T A95B returns text.

How do I use Qwen3.8 2.4T A95B through an API?

Qwen3.8 2.4T A95B is available through the Impossibl API. Use qwen/qwen3.8-2.4t-a95b 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 Qwen3.8 2.4T A95B released?

Qwen released Qwen3.8 2.4T A95B on August 12, 2026.

Creator release announcement ↗

this page describes Qwen3.8 2.4T A95B 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 →