Q
Qwen3.8 Max
The strongest Chinese closed model, first on APEX-Agents at 63.3%
Intelligence v5
34.0
Leaderboard rank
#14
Input / output
$2.00/$6.00
Blended 3:1
$3.00
Output speed
88 tok/s
Value
11 pts/$
Capability profile Normalised to the field maximum
Specifications
Vendor
Alibaba Qwen
Released
2026-09
Weights
Closed, API only
Context
1,000,000 tokens
Max output
Undisclosed
Parameters
Undisclosed
Input price
$2.00 / 1M
Output price
$6.00 / 1M
Cache hit
n/a
Output speed
88 tok/s
Intelligence
34.0（v5）
Benchmark scores
How it is measured →Ranked 14 · bars are scaled to the field maximum for that benchmark.
| Benchmarks | Score | Vs. best in field | Gap |
|---|---|---|---|
| AA Intelligence Index v5 Intelligence index | 34.0 | 57% | −26.0 pts |
| AA Intelligence Index v4.3.2 | — | — | not indexed |
| GPQA Diamond Graduate-level science reasoning | 92.7 | 96% | −3.7 pts |
| Terminal-Bench 4.0 Terminal-agent tasks | 38.9 | 58% | −28.3 pts |
| LiveBench Design Document generation (DOCX/PPT/PDF) | 78.6 | 90% | −8.3 pts |
| Vals Index | — | — | not indexed |
Assessment
An APEX-Agents score of 63.3% beats every self-hostable open model; Terminal-Bench 4.0 is 38.9%. Its open foundation, Qwen3.8, is on the open-weights table.
At the same price
The four models closest on blended price, side by side.
| Model | v5 | Blended | tok/s | Score gap |
|---|---|---|---|---|
| Grok 4.7xAI | 34.3 | $3.00 | 105 | +0.3 |
| GLM-5.3 Open weightsZ.ai | 32.0 | $2.15 | 77 | -2.0 |
| Claude Sonnet 5.5Anthropic | 54.5 | $4.00 | 135 | +20.5 |
| GPT-6.1 SolOpenAI | 54.6 | $4.00 | 77 | +20.6 |
Compare side by side
Line it up against other models on price, speed and every benchmark.
The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.
Open the comparison tool → Back to leaderboard