Qwen 3.8 Max
qwen/qwen3.8-max
autonomous long-horizon coding & professional work.
cmd --model qwen/qwen3.8-max
Intelligence index
40.3
Output speed
41.6 tok/s
Input
$2 /M
Output
$6 /M
Cache read
$0.25 /M
Agent-loop cost
$0.78 /M in
Context window
1M tokens
Released
August 3, 2026
Modalities
→
vs. the lineup
Qwen 3.8 Max beside its stablemates and nearest rivals. The ◆ marks the best value in each column across every row shown.
pin a rival:
| Model | Intelligence | Coding | Speed | Input $/M | Output $/M | Blended $/M | Context |
|---|---|---|---|---|---|---|---|
| Qwen 3.8 Max ◆ | 40.3◆ | 71.8◆ | 41.6◆ | $2◆ | $6◆ | $3◆ | 1M◆ |
| Qwen 3.8 27B | 33.9◆ | 68.1◆ | 42.7◆ | $0.40◆ | $3◆ | $1.05◆ | 262K◆ |
| Qwen 3.7 Max | 29.9◆ | 66◆ | —◆ | $2.50◆ | $7.50◆ | $3.75◆ | 1M◆ |
| Qwen 3.7 Plus | 25.8◆ | 55.9◆ | 71.6◆ | $0.40◆ | $1.60◆ | $0.70◆ | 1M◆ |
| Qwen 3.8 Max 0902 | —◆ | —◆ | —◆ | $2◆ | $6◆ | $3◆ | 1M◆ |
| Qwen 3.8 Flash | —◆ | —◆ | —◆ | $0.16◆ | $0.47◆ | $0.24◆ | 1M◆ |
| DeepSeek V4 Pro (latest)pinned | 36.3◆ | 68.8◆ | 65.3◆ | $0.66◆ | $1.98◆ | $0.99◆ | 1M◆ |
| DeepSeek V4 Flash (latest)pinned | 35◆ | 69.1◆ | 236.4◆ | $0.15◆ | $0.60◆ | $0.26◆ | 1M◆ |
coding performance
The Intelligence Index and its sub-scores, ranked against every scored model in the catalog. A metric that has not been measured for Qwen 3.8 Max has been left empty.
Coding Index
71.8
#20 of 53 scored
Terminal-Bench
81.3
#19 of 53 scored
Intelligence Index
40.3
#16 of 60 scored
Long-context reasoning
78.3
reasoning across a long context
SciCode
53.2
scientific coding
GPQA Diamond
92.7
graduate-level QA
usage calculator
How far a month of credits goes on Qwen 3.8 Max.
Input tokensfresh prompt
800Output tokensmodel reply
180Cache read tokensre-read context
50Kcost / request $0.015 · in $2 · out $6 · cache $0.25 per M
fresh input 11%output 7%cache reads 82%
Requests / 30 days
659
$10 credits ÷ $0.015 per request
~132 quick fixes~26 bug fixes~4 feature PRs
what real work costs
Real coding tasks priced end to end on Qwen 3.8 Max, from a quick lookup to a full-repo agent run.
One agent task
$0.20
180K in at 75% cache hit, 12K out
What you pay
$1.02 /M
all-in across every token that task touched
Sticker input
$2 /M
cache reads bill at $0.25 /M instead
| Task | Tokens in · out | Qwen 3.8 Max | Muse Spark 1.3 | Claude Haiku 4.5 |
|---|---|---|---|---|
| Quick lookup / one-liner | 8K · 1K | $0.0094 | $0.0063 | $0.0065 |
| Review a 500-line PR | 60K · 4K | $0.07 | $0.05 | $0.04 |
| Fix a bug (agent loop) | 180K · 12K | $0.20 | $0.13 | $0.12 |
| Refactor a module | 320K · 20K | $0.36 | $0.23 | $0.21 |
| Full-repo agent run | 900K · 45K | $0.84 | $0.54 | $0.49 |
frequently asked
How much does Qwen 3.8 Max cost?
$2/M input and $6/M output, cache reads $0.25/M. In an agent loop most input is cache-read, so the effective input rate is about $0.78/M.Which plan do I need?
Available on Go and above.
How do I switch to it?
Run
cmd --model qwen/qwen3.8-max, or type /model in a session and pick it. You can switch mid-session without losing context.Ship code that matches your taste
Command Code is the AI coding agent that continuously learns your taste. Start for $1.
Benchmarks from Artificial Analysis (v4.3)commandcode.ai/models/qwen3-8-max