AlibabaQwen3.8-Flash-Next
Best configuration ranks #108 of 610 on the BenchLeader Index at 59.3 ±7.4. Last measured 8 Sept 2026. Released 26 Aug 2026.
- Blended price
- $0.230/M
- $0.150 in · $0.470 out
- Output speed
- 64 tok/s
- First answer
- 34 s
- Context
- 256k
- Overall index59
- Coding71
- Agents & tools49
- Knowledge59
- Multimodal64
- Long context66
- Composite69
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LiveBench Reasoningnot in index | 87.4%#20 | LiveBench | |
| GPQA Diamond (AA)not in index | 92.3%#36 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 38.0%#66 | Artificial Analysis |
Coding
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LMArena WebDev | 1631#8 | LMArena | |
| LiveBench Codingnot in index | 72.5%#42 | LiveBench | |
| SciCode (AA)not in index | 50.6%#63 | Artificial Analysis |
Agents & tools
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LMArena Agent | 0.8#24 | LMArena | |
| LiveBench Agentic Codingnot in index | 61.6%#8 | LiveBench |
Maths
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LiveBench Mathematicsnot in index | 85.8%#39 | LiveBench |
Knowledge
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LiveBench Data Analysisnot in index | 74.2%#29 | LiveBench | |
| AA-Omniscience | -9.7#136 | Artificial Analysis |
Instruction following
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LiveBench Languagenot in index | 74.6%#40 | LiveBench |
Multimodal
| Benchmark | default | Source | Trend |
|---|---|---|---|
| MMMU-Pro | 79.8%#40 | Artificial Analysis |
Long context
| Benchmark | default | Source | Trend |
|---|---|---|---|
| AA-LCR | 79.7%#59 | Artificial Analysis |
Composite
| Benchmark | default | Source | Trend |
|---|---|---|---|
| LiveBench | 76.2%#22 | LiveBench | |
| AA Intelligence Index | 42.2#28 | Artificial Analysis |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | – | 38.9 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0021 | – | 43.6 s |
| Code edit | 6,000 / 1,500 | $0.0016 | – | 57.8 s |
| Agentic coding session | 60,000 / 4,000 | $0.011 | – | 1.6 min |
| Structured extraction | 2,000 / 200 | $0.0004 | – | 37.3 s |
See also
Data as of 9 Sept 2026. Compare with another model.