Qwen language model

Qwen4 Max

Qwen4 Max is Alibaba's flagship Qwen4 model: a hosted 2.4T-parameter MoE that takes text, image and video, with a thinking mode. it succeeds Qwen3.8 Max at lower rates, and its weights are not published.

01

price

$1.5/M

input tokens

$4.5/M

output tokens

$0.075/M

cached input

per 1M tokens. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).

  • cache hits on the implicit context cache bill at $0.075 per 1M; cache writes bill at the normal input rate.

02

what real work costs

workloadcost
one chat reply 3,000 tokens in, 500 out $0.0068
summarise a 100-page report 50,000 tokens in, 1,500 out, one request $0.08
1,000 classification calls 500 tokens in, 20 out, per call $0.84
coding-agent session 2M tokens in across ~60 turns, 60,000 out, no cache $3.27
same session, cached 85% of input read from the prompt cache, the rest written to it $0.85

at the listed rate. token counts are typical sizes, not measurements.

03

specs

provider
Qwen
context window
1M tokens — ≈ 1,500 pages of prose or 100,000 lines of code
max output
262k tokens
input
text, images, video
output
text
structured output
yes — JSON schema
reasoning mode
yes
prompt caching
yes — cache hits at $0.075/M
weights
closed
released
November 2026
status
live
model id
qwen4-max

integration notes

  • max input is 991,808 tokens (983,616 in thinking mode), below the context window; max output is 131,072 tokens without thinking and 262,144 with it.
  • preserve_thinking is on by default: send earlier turns' reasoning back unchanged with each request.
  • reasoning effort takes low, high or xhigh and defaults to high; medium is mapped to high and max to xhigh.

04

benchmarks

as published by the maker; not measured by octomind.

benchmarkscore
DeepSWE v1.1source75.7
Terminal-Bench 4.0source48.5
HealthBench Professionalsource60.0
Humanity's Last Examsource51.7
Work360source52.69

05

when to pick it

pick it if

  • you want Qwen4's flagship with image and video input
  • long agent sessions reuse a prefix; cache hits bill $0.075 per 1M, against $0.25 on Qwen3.8 Max
  • token-heavy agent runs; Alibaba reports 48 to 49% fewer tokens than Qwen3.8 Max on Terminal-Bench 4.0

06

where it sits on price

blended price $2.25/M at 3 input tokens per output token: 30 live language models cost less, 13 cost more.

  1. Mistral Large 4 $2.065/M
  2. GLM-5.3 $2.15/M
  3. Qwen4 Max $2.25/M
  4. Grok 4.7 $3/M
  5. Qwen3.8 Max $3/M

07

use Qwen4 Max in octomind

one octohub login reaches every model in the catalog. pick this one per run, per session, or as your default.

run it
octomind login
octomind run -m octohub:qwen4-max
switch mid-session
/model octohub:qwen4-max
make it the default — config.toml
[model]
name = "octohub:qwen4-max"

08

compare Qwen4 Max

all model comparisons →

09

similar models

nearest live models by listed input price.

view all models →

10

earlier versions

retired models retained for historical reference.

11

Qwen4 Max faq

how much does Qwen4 Max cost?

Qwen4 Max lists at $1.5/M input, $4.5/M output, $0.075/M cached input per 1M tokens. a typical coding-agent session (2M tokens in, 60,000 out) comes to $3.27, or $0.85 when 85% of the input is read from the prompt cache. octomind bills the listed provider price, with no markup, on every plan, including the free one (pay as you go from credits).

what is the context window of Qwen4 Max?

1M tokens, ≈ 1,500 pages of prose or 100,000 lines of code. a single reply can run to 262k tokens.

does Qwen4 Max accept images?

yes. it takes text, images and video as input and returns text.

is Qwen4 Max open source?

no. Qwen4 Max is a closed-weights model, available only through Qwen's api and hosted gateways such as octohub.

when was Qwen4 Max released?

November 2026, by Qwen.

is Qwen4 Max free?

no. Qwen4 Max bills per token at $1.5/M input, $4.5/M output, $0.075/M cached input per 1M. octomind's free plan reaches it pay as you go from credits, so you pay only for the tokens you use, at the provider price. credits carry a 3% top-up fee.

how do i use Qwen4 Max in octomind?

run octomind login once, then octomind run -m octohub:qwen4-max. inside a running session, /model octohub:qwen4-max switches to it without losing context.

use Qwen4 Max in octomind

choose a model when you start, then switch mid-session when the work changes.