Qwen3.7 Max vs Qwen3.8 Max: Specs & Benchmark Comparison

Qwen3.7 Max and Qwen3.8 Max both come from Alibaba, Qwen3.7 Max was released in May 2026, and Qwen3.8 Max followed 3 months later in August 2026. Qwen3.8 Max has a published size of about 2.4 trillion parameters; Alibaba has not disclosed the parameter count of Qwen3.7 Max.

The two models share 8 published benchmarks. Qwen3.8 Max leads on 8 of them. The widest gaps are on SkillsBench, where Qwen3.8 Max scores 70.2% against 59.2%; NL2Repo, where Qwen3.8 Max scores 55.9% against 47.2%. Averaged across everything we track, Qwen3.7 Max sits at 73.6% and Qwen3.8 Max at 71.0%.

On price the two are close: Qwen3.7 Max costs $2.50 per million input tokens and $7.50 per million output tokens, Qwen3.8 Max $2.50 and $6.25. Both accept a context window of about 1M tokens. Qwen3.7 Max accepts text as input, while Qwen3.8 Max accepts text and images.

CharacteristicQwen3.7 MaxQwen3.8 Max
CompanyAlibabaAlibaba
Release DateMay 19, 2026August 2, 2026
Parameters2.4T
MultimodalNoYes
Context (input)1.0M1.0M
Context (output)66K131K
Input Price / 1M$2.50$2.50
Output Price / 1M$7.50$6.25
Average Score73.6%71.0%
Benchmarks
SkillsBench59.2%70.2%
NL2Repo47.2%55.9%
CoWorkBench67.2%74.8%
SWE-Bench Pro60.6%67.7%
QwenSVG80.4%85.7%
IFBench79.1%82.8%
Humanity's Last Exam41.4%43.6%
GPQA92.4%92.6%

Visual Benchmark Comparison

Qwen3.7 Max
Qwen3.8 Max
SkillsBench0.6 vs 0.7
0.6
0.7
NL2Repo0.5 vs 0.6
0.5
0.6
CoWorkBench0.7 vs 0.7
0.7
0.7
SWE-Bench Pro0.6 vs 0.7
0.6
0.7
QwenSVG0.8 vs 0.9
0.8
0.9
IFBench0.8 vs 0.8
0.8
0.8
Humanity's Last Exam0.4 vs 0.4
0.4
0.4
GPQA0.9 vs 0.9
0.9
0.9

Verdict

Qwen3.8 Max leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: Qwen3.7 Max — 0.7, Qwen3.8 Max — 0.7.

API Cost

Qwen3.8 Max is 1.1x cheaper: input $2.50/1M vs $2.50/1M tokens.

Context Window

Same context size: 1M tokens.

Recency

Qwen3.8 Max is newer: released 8/2/2026 vs 5/19/2026.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — Qwen3.7 Max or Qwen3.8 Max?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — Qwen3.7 Max or Qwen3.8 Max?
Qwen3.7 Max is cheaper for input: $2.50 per 1M tokens vs $2.50.
Which has a larger context window — Qwen3.7 Max or Qwen3.8 Max?
Qwen3.7 Max supports a larger context: 1,000,000 tokens vs 1,000,000.

The Qwen3.7 Max and Qwen3.8 Max comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Qwen3.7 Max or Qwen3.8 Max page. See also the complete list of AI model comparisons.