AlibabaQwen3.8-Flash-Next
Qwen3.8-Flash-Next is an Alibaba open-weights reasoning model, released 26 Aug 2026. Its best configuration ranks #114 of 751 on the BenchLeader Index at 59.9 ±6.0, in the upper half. The ± is the point: 243 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in coding (68) and lowest in knowledge (58). At $0.230 per million tokens blended it is cheaper than most ranked models. Output speed of 54 tokens per second puts it slower than most, with a first answer in 39.7 s. Last measured 2 Oct 2026.
- Blended price
- $0.230/M
- $0.150 in · $0.470 out
- Output speed
- 54 tok/s
- measured by Artificial Analysis
- First answer
- 40 s
- Context
- 256k
- Full answer
- 49 s
- median, reasoning included
- Cost per run
- $0.372
- one full Intelligence Index run
- Released
- 26 Aug 2026
- Overall index60
- Reasoning60
- Coding68
- Agents & tools65
- Knowledge58
- Multimodal63
- Long context65
- Composite64
How its index has moved
9 Sept 2026 to 3 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
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 | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Reasoningnot in index | 87.4%#26 | LiveBench | |
| GPQA Diamond (AA)not in index | 92.3%#36 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 38.0%#93 | Artificial Analysis | |
| CritPt | 11.1%#100 | Artificial Analysis |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LMArena WebDev | 1638#15 | LMArena | |
| LiveBench Codingnot in index | 72.5%#52 | LiveBench | |
| SciCode (AA)not in index | 50.6%#92 | Artificial Analysis | |
| Terminal-Bench 4.0 (AA)not in index | 25.3%#46 | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 86.1%#21 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LMArena Agent | -0.7#33 | LMArena | |
| LiveBench Agentic Codingnot in index | 61.6%#11 | LiveBench | |
| GDPval (AA)not in index | 55.6%#20 | Artificial Analysis | |
| τ²-Bench Banking (AA) | 45.4%#14 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Mathematicsnot in index | 85.8%#49 | LiveBench |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Data Analysisnot in index | 74.2%#37 | LiveBench | |
| AA-Omniscience | -9.7#173 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 24.5%#223 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 54.7%#88 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Languagenot in index | 74.6%#49 | LiveBench | |
| LiveBench Instruction Followingnot in index | 77.1%#5 | LiveBench |
Multimodal
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| MMMU-Pro | 79.8%#54 | Artificial Analysis |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 79.7%#80 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench | 76.2%#30 | LiveBench | |
| AA Intelligence Index | 39.8#56 | 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 | – | 45.3 s |
| Summarise a 30-page report | 12,000 / 600 | $0.0021 | – | 50.9 s |
| Code edit | 6,000 / 1,500 | $0.0016 | – | 1.1 min |
| Agentic coding session | 60,000 / 4,000 | $0.011 | – | 1.9 min |
| Structured extraction | 2,000 / 200 | $0.0004 | – | 43.4 s |
See also
Data as of 4 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Qwen3.8-Flash-Next: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/qwen3-8-flash-next, data as of 4 Oct 2026.