Which is cheaper, Grok 4.7 or Qwen3.8-Max?

Per token, neither: both charge $2 per million tokens read and $6 per million written. The same text can still be a different number of tokens on each, so the bills can differ.

Which is better, Grok 4.7 or Qwen3.8-Max?

On the Epoch Capabilities Index, an independent score from Epoch AI that combines over 50 benchmarks, Qwen3.8-Max scores 157 and Grok 4.7 154, above the top of the range of doubt around Grok 4.7's score (GPT-5 is 150). One score is an average; test both on your own work before you switch.

What does a support bot cost a month on Grok 4.7 and Qwen3.8-Max?

At 2,000 replies a day, each reading 4,000 tokens (50 percent of them from cache) and writing 300, it costs $408 a month on Grok 4.7 and $378 on Qwen3.8-Max: Grok 4.7 costs 1.1 times as much for this job.

What does an AI agent task cost on Grok 4.7 and Qwen3.8-Max?

One 40-turn task that reads 2,800,000 tokens in all, 90 percent of them from cache, and writes 40,000, costs $2.06 on Grok 4.7 and $1.43 on Qwen3.8-Max. At 10 tasks a day that is $618 and $429 a month.

Does the same text take the same number of tokens on Grok 4.7 and Qwen3.8-Max?

No. Our lesson "How AI writes", 1,700 words long, is about 2,100 tokens on Grok 4.7 and about 2,100 on Qwen3.8-Max. Each maker cuts text its own way, and every extra token is paid for at the model's price.

Can I pay less on Grok 4.7 and Qwen3.8-Max for work that can wait?

Not with a batch price: neither maker's price page lists one for these models. A cache still lowers the price of text the model reads again.

Grok 4.7 vs Qwen3.8-Max

Grok 4.7 and Qwen3.8-Max charge the same per token: $2 per million tokens read and $6 per million written. The bill can still differ, because the same text can be a different number of tokens on each. Below are both price lists, 8 everyday jobs priced on each, and the count for one text.

Price your own job Compare 2 other models

For the same tokens: The same price, $2 in and $6 out
Cheaper to read: Neither, both $2
Cheaper to write: Neither, both $6
Scores higher: Qwen3.8-Max, 157 against 154
Larger window: Not listed for Qwen3.8-Max

Price per million tokens, side by side

In US dollars, as each maker prints them. The better figure in each row is green.

	Grok 4.7xAI	Qwen3.8-MaxAlibaba
InputTokens the model reads	$2	$2
Cached inputRead again from the cache	$0.50	$0.25
Cache writePutting text in the cache	Not listed	Not listed
OutputTokens it writes, thinking included	$6	$6
Blend 3:13 tokens read for each 1 written	$3	$3
Batch inputAnswers collected later	Not listed	Not listed
Batch output	Not listed	Not listed
Long prompt inputAbove 200K on Grok 4.7	$4	$2
Long prompt outputThe standard price where there is no long-prompt price	$12	$6
Context windowTokens it holds at once	500K	Not listed
Capability scoreEpoch Capabilities Index; GPT-5 is 150	154	157
Longest answerTokens it writes in 1 reply	Not listed	Not listed

The same text, counted by each maker

A price per token means little until you know how many tokens your text is. Here is our lesson How AI writes, about 1,700 words of English, counted the way each maker counts it, and what it costs to send it once.

Grok 4.7
about 2,100 tokens
Sent once: 0.41 cents at $2 per million
xAI publishes neither its tokenizer nor a rule for estimating. The figure is OpenAI's o200k_base count, a rough guide only.
Qwen3.8-Max
about 2,100 tokens
Sent once: 0.43 cents at $2 per million
Counted with the tokenizer Alibaba publishes with its open Qwen3.8 models. Alibaba does not say its API models use the same one, so the count is marked as about.

What 8 everyday jobs cost a month on each

Each job at its usual number of runs a day, for 30 days, at the standard price. The cheaper of the 2 is green; open any job in the calculator to change it.

Job	Grok 4.7	Qwen3.8-Max	Difference
A chat question300 in, 500 out, 100 a day	$10.800.36 cents a run	$10.800.36 cents a run	The same
A hard question, with thinking300 in, 1,400 out, 100 a day	$27.000.9 cents a run	$27.000.9 cents a run	The same
A long report, summarised60,000 in, 2,000 out, 20 a day	$79.2013 cents a run	$79.2013 cents a run	The same
A 40-turn agent task2,800,000 in, 40,000 out, 10 a day	$618$2.06 a run	$429$1.43 a run	1.4 times
A support bot's reply4,000 in, 300 out, 2,000 a day	$4080.68 cents a run	$3780.63 cents a run	1.1 times
Sort a message into a category250 in, 5 out, 50,000 a day	$7950.053 cents a run	$7950.053 cents a run	The same
Read an invoice into fields1,500 in, 250 out, 1,000 a day	$1350.45 cents a run	$1350.45 cents a run	The same
Review a pull request25,000 in, 4,500 out, 50 a day	$1167.7 cents a run	$1167.7 cents a run	The same

Your own job on both

Start from a job, then set its size, how often it runs and whether the answers can wait. Both bills follow every change.

Start from a job

A short question and a few paragraphs back.

Read in 1 run

tokens

about 230 words

Read again from cache

Instructions or documents the model has read before cost far less the second time.

Written in 1 run, thinking included

tokens

about 380 words

Runs a day

A month of this job

Grok 4.7$10.80
0.36 cents a run
Qwen3.8-Max$10.80
0.36 cents a run

The 2 cost the same for this job, counted in tokens.

This job on every model

Why the price per token is only the start

Every maker prints 2 main prices: one for the tokens a model reads (input) and one for the tokens it writes (output). On most current models writing costs 3 to 5 times as much as reading, so which model is cheaper can depend on the job. A job that reads a long document and writes a line pays mostly the input price; a chat that writes paragraphs pays mostly the output price. That is why each comparison prices 8 jobs and not one figure.

The same text is not the same number of tokens on every model. Each maker cuts text its own way, and Anthropic says its models from Claude 4.7 on produce about 30 percent more tokens for the same text than earlier ones. A model with the lower price per token can still send the larger bill, so each comparison counts one of our lessons the way both makers count it.

Discounts change the order too. Work that can wait costs half on a batch at the makers that offer one, and text read again from a cache costs a quarter of the input price or less on every current model here. The price table on each comparison shows who offers what.

Thinking is billed as output, and models differ in how much they think and how long they answer. A price list cannot show that. Before you switch, run the same 10 tasks on both models and compare the bills.

Questions

Which is cheaper, Grok 4.7 or Qwen3.8-Max?: Per token, neither: both charge $2 per million tokens read and $6 per million written. The same text can still be a different number of tokens on each, so the bills can differ.
Which is better, Grok 4.7 or Qwen3.8-Max?: On the Epoch Capabilities Index, an independent score from Epoch AI that combines over 50 benchmarks, Qwen3.8-Max scores 157 and Grok 4.7 154, above the top of the range of doubt around Grok 4.7's score (GPT-5 is 150). One score is an average; test both on your own work before you switch.
What does a support bot cost a month on Grok 4.7 and Qwen3.8-Max?: At 2,000 replies a day, each reading 4,000 tokens (50 percent of them from cache) and writing 300, it costs $408 a month on Grok 4.7 and $378 on Qwen3.8-Max: Grok 4.7 costs 1.1 times as much for this job.
What does an AI agent task cost on Grok 4.7 and Qwen3.8-Max?: One 40-turn task that reads 2,800,000 tokens in all, 90 percent of them from cache, and writes 40,000, costs $2.06 on Grok 4.7 and $1.43 on Qwen3.8-Max. At 10 tasks a day that is $618 and $429 a month.
Does the same text take the same number of tokens on Grok 4.7 and Qwen3.8-Max?: No. Our lesson "How AI writes", 1,700 words long, is about 2,100 tokens on Grok 4.7 and about 2,100 on Qwen3.8-Max. Each maker cuts text its own way, and every extra token is paid for at the model's price.
Can I pay less on Grok 4.7 and Qwen3.8-Max for work that can wait?: Not with a batch price: neither maker's price page lists one for these models. A cache still lowers the price of text the model reads again.

Sources