GPQA DiamondVerified
93.5%
GPQA Diamond (AA run)
Models / Alibaba
Qwen3.8 2.4T A95B
Qwen · Released 2026-08-12
Alibaba's open-weight Qwen-Max-class MoE (12 Aug 2026): 2.4T total / 95B active. AA list price $2 / $6 per 1M tokens.
Self-hostable Qwen3.8 checkpoint for long-context and agentic work. Hosted Qwen3.8 Max adds vision and a non-thinking mode. Prefer sourced ledger rows over vendor eval tables.
Composite70
Context262K
Input / 1M$2.0
Output / 1M$6.0
Knowledge cutoffunpublished
Output speed37 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
78/100 avg
- Reasoning
- 94
- Coding
- 54
- Agentic
- 82
- Long context
- 80
4 tested benchmark families
Editor's note
This is the open-weight Max-class checkpoint, not the hosted Qwen3.8 Max API product. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where Qwen3.8 2.4T A95B 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 |
|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | #13/ 184 | 94 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench | Agentic | #20/ 167 | 82 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #29/ 97 | 54 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #32/ 186 | 42 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #37/ 184 | 80 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #138/ 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 2.4T A95B shows up, it's highlighted.
ReasoningQwen3.8 2.4T A95B · 94
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingQwen3.8 2.4T A95B · 54
- 174
- 273
- 370
AgenticQwen3.8 2.4T A95B · 82
- 191
- 289
- 389
Long contextQwen3.8 2.4T A95B · 80
- 189
- 286
- 385
Newest receipts
6 recentLast 120 days
Benchmark rows added to the public ledger for Qwen3.8 2.4T A95B in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 37 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 80.33% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 93.5% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 42.4% | Artificial Analysis | 24 |
| SciCode | Coding | 54.1% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 82.02% | Artificial Analysis | 24 |
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 2.4T A95B 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 2.4T A95B. Scores link to the original source; gaps mean no public row exists yet.
6 of 13 catalog benchmarks have a sourced row for Qwen3.8 2.4T A95B.
Reasoning
Humanity's Last ExamVerified
42.4%
HLE (AA run)
Coding
SciCodeVerified
54.1%
SciCode (AA run)
Agentic
Terminal-BenchVerified
82.02%
Terminal-Bench v2.1 (AA run)
Long context
AA-LCRVerified
80.33%
AA-LCR (AA run)
Performance
AA output speedVerified
37 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Open weights published.
Sources
Where the numbers come from
- Hugging Face Qwen3.8-2.4T-A95B· checked 2026-08-15
- Artificial Analysis Qwen3.8 2.4T A95B· checked 2026-08-15
- Alibaba Cloud Qwen3.8-Max blog· checked 2026-08-15
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 2.4T A95B?
Qwen3.8 2.4T A95B is a Alibaba model: Alibaba's open-weight Qwen-Max-class MoE (12 Aug 2026): 2.4T total / 95B active. AA list price $2 / $6 per 1M tokens.
What is Qwen3.8 2.4T A95B for?
Self-hostable Qwen3.8 checkpoint for long-context and agentic work. Hosted Qwen3.8 Max adds vision and a non-thinking mode. Prefer sourced ledger rows over vendor eval tables.
What does Qwen3.8 2.4T A95B cost?
Qwen3.8 2.4T A95B lists $2.0 input and $6.0 output per 1M tokens. Pricing last checked 2026-08-15.
How large is the Qwen3.8 2.4T A95B context window?
Qwen3.8 2.4T A95B publishes a 262K token context window.
What is the VerdictPal status of Qwen3.8 2.4T A95B?
Qwen3.8 2.4T A95B carries Solid 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.