Alibaba Cloud / Qwen Team model product
Qwen3.8 Flash Next
8 Flash Next is an open-weight experimental preview of the architecture planned for Qwen4.
Updated Sep 8, 2026. Default version: Qwen3.8-Flash-Next
LLMBoard Score81.4Qwen3.8-Flash-Next
Coverage40%22 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable
Qwen3.8 Flash Next Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen3.8 Flash Next Benchmark Results
Benchmark scores for Qwen3.8-Flash-Next.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkClawEval-MM | Score60.40% | Rank01 | Participants6 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score91.90% | Rank01 | Participants62 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkRealWorldQA | Score88.50% | Rank01 | Participants31 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkRecreationBench | Score49.90% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAndroidWorld | Score84.50% | Rank02 | Participants7 | Percentile83.33% | EvidenceC | Evaluated |
| BenchmarkCoWorkBench | Score73.90% | Rank02 | Participants6 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkERQA | Score72.30% | Rank02 | Participants26 | Percentile96.00% | EvidenceC | Evaluated |
| BenchmarkMathVision | Score95.70% | Rank02 | Participants35 | Percentile97.06% | EvidenceC | Evaluated |
| BenchmarkVision2Web | Score64.00% | Rank02 | Participants5 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkJob Bench | Score55.70% | Rank03 | Participants8 | Percentile71.43% | EvidenceC | Evaluated |
| BenchmarkAgents' Last Exam | Score51.20% | Rank04 | Participants16 | Percentile80.00% | EvidenceC | Evaluated |
| BenchmarkCharXiv-R | Score90.60% | Rank04 | Participants55 | Percentile94.44% | EvidenceC | Evaluated |
| BenchmarkIFBench | Score81.30% | Rank04 | Participants39 | Percentile92.11% | EvidenceC | Evaluated |
| BenchmarkSWE-bench Multilingual | Score81.00% | Rank04 | Participants43 | Percentile92.86% | EvidenceC | Evaluated |
| BenchmarkToolathlon | Score73.50% | Rank05 | Participants41 | Percentile90.00% | EvidenceC | Evaluated |
| BenchmarkLVBench | Score76.60% | Rank06 | Participants28 | Percentile81.48% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score12.46% | Rank06 | Participants49 | Percentile89.58% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,626.46 rating | Rank07 | Participants97 | Percentile93.75% | EvidenceA | Evaluated |
| BenchmarkNL2Repo | Score48.10% | Rank09 | Participants22 | Percentile61.90% | EvidenceC | Evaluated |
| BenchmarkOSWorld 2.0 | Score19.40% | Rank10 | Participants11 | Percentile10.00% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Pro | Score62.50% | Rank12 | Participants55 | Percentile79.63% | EvidenceC | Evaluated |
| BenchmarkDeepSWE 1.1 | Score58.70% | Rank17 | Participants31 | Percentile46.67% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score91.70% | Rank19 | Participants247 | Percentile92.68% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score2.61% | Rank23 | Participants49 | Percentile54.17% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score0.24% | Rank27 | Participants49 | Percentile45.83% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score1.90% | Rank29 | Participants49 | Percentile41.67% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-1.18% | Rank29 | Participants49 | Percentile41.67% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score-0.39% | Rank42 | Participants49 | Percentile14.58% | EvidenceA | Evaluated |
| BenchmarkHumanity's Last Exam | Score35.90% | Rank45 | Participants103 | Percentile56.86% | EvidenceC | Evaluated |
Qwen3.8 Flash Next Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent task outcome explicit | Categoryoverall | Rank06 | Rating / score0.12 | VotesN/A | Observations9.5K | Result date |
| Arenawebdev | Categorywebdev-react | Rank06 | Rating / score1,645.32 | Votes1,443 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank07 | Rating / score1,626.46 | Votes2,158 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank07 | Rating / score1,626.46 | Votes2,158 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank11 | Rating / score1,606.29 | Votes261 | ObservationsN/A | Result date |
| Arenaagent | Categoryoverall | Rank23 | Rating / score0.03 | VotesN/A | Observations2.9M | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank27 | Rating / score0.00 | VotesN/A | Observations3.4K | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank29 | Rating / score0.02 | VotesN/A | Observations126.7K | Result date |
| Arenaagent steerability | Categoryoverall | Rank29 | Rating / score-0.01 | VotesN/A | Observations8.2K | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank42 | Rating / score0.00 | VotesN/A | Observations2.7M | Result date |
Qwen3.8 Flash Next Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.12 input, $0.40 output per 1M via Vercel AI Gateway
- Tracked offerings
- 4
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderVercel AI Gateway | Provider model IDalibaba/qwen3.8-flash-next | Regionglobal | Input / 1M$0.12 | Output / 1M$0.40 | Context1M | Updated |
| ProviderAMD | Provider model IDQwen3.8-Flash-Next | Regionglobal | Input / 1M$0.15 | Output / 1M$0.47 | Context262.1K | Updated |
| ProviderRequesty | Provider model IDqwen3.8-flash-next | Regionglobal | Input / 1M$0.20 | Output / 1M$0.50 | Context262.1K | Updated |
| ProviderCortecs | Provider model IDqwen3.8-flash-next | Regionglobal | Input / 1M$0.201 | Output / 1M$0.50 | Context262.1K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen3.8 Flash Next Runtime Performance
Provider-specific output speed and catalog latency for Qwen3.8-Flash-Next. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Qwen3.8 Flash Next Specifications
Technical details for the model's default version.
- Version
- Qwen3.8-Flash-Next
- Released
- Aug 26, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 125B
- Context window
- N/A
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Qwen Community License 1.0
Qwen3.8 Flash Next Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionQwen3.8-Flash-Next | Released | LLMBoard81.44 | Parameters125B | ContextN/A | Max outputN/A | Open weightsYes | LicenseQwen Community License 1.0 |
Models similar to Qwen3.8 Flash Next
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen3.8 Flash
Alibaba Cloud / Qwen Team
AC
Qwen3.7 Max
Alibaba Cloud / Qwen Team
AC
Qwen3.8 Max
Alibaba Cloud / Qwen Team
AC
Qwen3.8 27B
Alibaba Cloud / Qwen Team
AC
Qwen3.7 Plus
Alibaba Cloud / Qwen Team
AC
Qwen3.6 Plus
Alibaba Cloud / Qwen Team
What is Qwen3.8 Flash Next?
Key information about Qwen3.8 Flash Next and its available data.
8 Flash Next is a causal language model from Alibaba Cloud / Qwen Team with a vision encoder for text, image, and video, Hybrid Attention (QSA), Gated Residual, and N-gram Embedding. It supports a native 262,144-token context that can be extended to 1,000,000 tokens and has thinking enabled by default.
Data as of 2026-09-08.
FAQ
Common questions about Qwen3.8 Flash Next.
When was Qwen3.8 Flash Next released?
Qwen3.8 Flash Next's default version was released on Aug 26, 2026.
How much does Qwen3.8 Flash Next cost?
No official standard PAYG price is currently available for Qwen3.8 Flash Next. The lowest tracked third-party offer starts at $0.12 input and $0.40 output via Vercel AI Gateway.
Who created Qwen3.8 Flash Next?
Qwen3.8 Flash Next was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen3.8 Flash Next?
A context window is not available for the default version.
Is Qwen3.8 Flash Next open weight?
Yes. The default version is marked as open weight under Qwen Community License 1.0.