GPQA DiamondVerified
92.8%
GPQA Diamond (AA run)
Models / Alibaba
Qwen3.8 Max
Qwen · Released 2026-08-03
Alibaba's August 2026 Max-tier flagship. API list $2 / $6 per 1M tokens on that snapshot.
Current Alibaba Cloud Max model for long-context and agentic API work. Prefer sourced ledger rows over vendor eval tables; AA scores sync from the public snapshot.
Composite72
Context1M
Input / 1M$2.0
Output / 1M$6.0
Knowledge cutoffunpublished
Output speed36 t/s
Statusflagship
Family profile
Best published score in each covered benchmark family.
79/100 avg
- Reasoning
- 93
- Coding
- 52
- Agentic
- 89
- Multimodal
- 82
- Long context
- 80
5 tested benchmark families
Editor's note
Recorded from the Alibaba Cloud launch plus AA snapshots. Editorial depth still needs a desk pass. Vendor PaperBench/OSWorld figures stay off the ledger.Benchmark placements
Per-benchmark positions
Where Qwen3.8 Max places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| Terminal-Bench | Agentic | #3/ 167 | 89 | Artificial Analysis | 2026-09-28 |
| MMMU-Pro | Multimodal | #6/ 8 | 82 | Artificial Analysis — MMMU-Pro | 2026-09-01 |
| GPQA Diamond | Reasoning | #20/ 184 | 93 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #25/ 186 | 43 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #36/ 97 | 52 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #38/ 184 | 80 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #140/ 195 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Qwen3.8 Max shows up, it's highlighted.
ReasoningQwen3.8 Max · 93
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingQwen3.8 Max · 52
- 174
- 273
- 370
AgenticQwen3.8 Max · 89
- 191
- 289
- 389
MultimodalQwen3.8 Max · 82
- 186
- 285
- 384
Long contextQwen3.8 Max · 80
- 189
- 286
- 385
Newest receipts
7 recentLast 120 days
Benchmark rows added to the public ledger for Qwen3.8 Max in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 36 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 80.33% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 92.8% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 43.1% | Artificial Analysis | 24 |
| SciCode | Coding | 52.1% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 88.76% | Artificial Analysis | 24 |
| MMMU-Pro | Multimodal | 82.3% | Artificial Analysis — MMMU-Pro | 51 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Qwen3.8 Max sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Qwen3.8 Max. Scores link to the original source; gaps mean no public row exists yet.
7 of 13 catalog benchmarks have a sourced row for Qwen3.8 Max.
Reasoning
Humanity's Last ExamVerified
43.1%
HLE (AA run)
Coding
SciCodeVerified
52.1%
SciCode (AA run)
Agentic
Terminal-BenchVerified
88.76%
Terminal-Bench v2.1 (AA run)
Multimodal
MMMU-ProVerified
82.3%
MMMU-Pro (AA run)
Long context
AA-LCRVerified
80.33%
AA-LCR (AA run)
Performance
AA output speedVerified
36 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Dated 2 Sep snapshot remains on Model Studio. International list unchanged.
- List price unchanged at $2 / $6.
Sources
Where the numbers come from
- Alibaba Cloud Qwen3.8-Max blog· checked 2026-08-12
- Alibaba Cloud Model Studio· checked 2026-08-12
- Alibaba Model Studio pricing· checked 2026-09-28
- Artificial Analysis Qwen3.8 Max· checked 2026-09-10
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Qwen3.8 Max?
Qwen3.8 Max is a Alibaba model: Alibaba's August 2026 Max-tier flagship. API list $2 / $6 per 1M tokens on that snapshot.
What is Qwen3.8 Max for?
Current Alibaba Cloud Max model for long-context and agentic API work. Prefer sourced ledger rows over vendor eval tables; AA scores sync from the public snapshot.
What does Qwen3.8 Max cost?
Qwen3.8 Max lists $2.0 input and $6.0 output per 1M tokens. Pricing last checked 2026-09-28.
How large is the Qwen3.8 Max context window?
Qwen3.8 Max publishes a 1M token context window.
What is the VerdictPal status of Qwen3.8 Max?
Qwen3.8 Max carries Flagship status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.