Qwen3.8 Flash Next benchmarks
Qwen3.8 Flash Next is an open-weights model from Alibaba, released 26 Aug 2026. It ranks 25th of 51 on our overall leaderboard and 3rd of 50 for value, though some of its results are not yet published. At A$0.32 per million tokens (blended), it costs about 11× cheaper than the median model we track. It writes about 53 tokens a second, and answers start after 39 s on average, thinking included. Its weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region.
- CONTEXT WINDOW
- 262,144 tokens
- INPUT
- text, image, video
- WEIGHTS
- Open (qwen-community-1.0)
- RELEASED
- 26 Aug 2026
- OUTPUT SPEED
- 53 tok/s
- ANSWER STARTS AFTER
- 39 s (thinking included)
Qwen3.8 Flash Next across our rankings.
Benchmark results.
Independent test scores are for the default setting. “Among tracked” ranks Qwen3.8 Flash Next against the 51 models on this site; LMArena ranks run across its full leaderboard.
| BENCHMARK | RESULT | AMONG TRACKED | SOURCE DETAIL |
|---|---|---|---|
| INDEPENDENT TESTS · ARTIFICIAL ANALYSIS | |||
| AA Intelligence Index | 39.8 | #21 of 51 | Default |
| AA Coding Index | 73.1 | #17 of 42 | Default |
| GPQA Diamond | 92.3% | #19 of 44 | Default |
| Humanity’s Last Exam | 38.0% | #42 of 51 | Default |
| AA-LCR | 79.7% | #30 of 51 | Default |
| Terminal-Bench 2.1 | 86.1% | #9 of 42 | Default |
| τ²-Bench (banking) | 45.4% | #9 of 41 | Default |
| SciCode | 50.6% | #36 of 46 | Default |
| BLIND HUMAN VOTES · LMARENA | |||
| LMArena WebDev4,313 votes · ±10 | 1636 | #7 of 45 | #9 of 129 on LMArena · 22 Sep 2026 |
| LMArena Agent57,458 sessions | #26 | #23 of 37 | #26 of 46 on LMArena · 15 Sep 2026 |
What Qwen3.8 Flash Next costs in Australian dollars.
List prices converted at the snapshot’s RBA rate, excluding GST. Reasoning models are billed for their thinking as output tokens, so heavier settings cost more than the job estimates show.
| PER MILLION TOKENS | AUD | USD LIST |
|---|---|---|
| Input tokens | A$0.21 | US$0.15 |
| Output tokens | A$0.66 | US$0.47 |
| Blended (3 in : 1 out) | A$0.32 | US$0.23 |
| JOB | COST (AUD) | MEDIAN MODEL |
|---|---|---|
| Customer support replyPER 1,000 REPLIES | A$0.62 | ≈ A$6.58 |
| Summarise a 30-page documentPER 100 DOCUMENTS | A$0.47 | ≈ A$4.74 |
| Agentic coding taskPER 10 TASKS | A$0.37 | ≈ A$3.72 |
Running Qwen3.8 Flash Next in Australia.
Its weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region. How the platforms compare →
| PLATFORM | AVAILABILITY | DETAIL |
|---|---|---|
| AWS Bedrock · Sydney | NOT OFFERED | |
| AWS Bedrock · Melbourne | NOT OFFERED | |
| Azure · Australia East | NOT OFFERED | |
| Google Vertex AI · Sydney | NOT OFFERED | |
| Your own infrastructure | SELF-HOST | Open weights: run it on your own servers or GPU instances in an Australian region. |
Compare it with.
Qwen3.8 Flash Next, answered.
How much does Qwen3.8 Flash Next cost in Australian dollars?
Qwen3.8 Flash Next lists at US$0.15 per million input tokens and US$0.47 per million output tokens — A$0.21 and A$0.66 at A$1 = US$0.7123 (RBA, 22 Sep 2026), excluding GST. A typical customer support reply works out at about A$0.62 per 1,000 replies, before any reasoning tokens.
Can I use Qwen3.8 Flash Next in Australia with data kept onshore?
Qwen3.8 Flash Next’s weights are public, so it can be hosted in Australia on your own infrastructure or an Australian cloud region. Availability comes from the cloud providers’ own documentation; check your provider’s current terms before relying on it for data-residency obligations.
Is Qwen3.8 Flash Next good for coding?
It ranks 13th of 42 on our coding ranking, with an Artificial Analysis Coding Index of 73.1 and an LMArena WebDev rating of 1636 (9th of 129). The current leader is Claude Fable 5.1.
How good is Qwen3.8 Flash Next at agent and automation work?
It ranks 13th of 41 on our agents ranking, completing 45.4% of τ²-Bench banking customer-service tasks and 86.1% of Terminal-Bench 2.1 tasks.
How fast is Qwen3.8 Flash Next?
Artificial Analysis measures Qwen3.8 Flash Next at about 53 tok/s of output, with the answer starting after 39 s on average once thinking time is included (default setting).
What is Qwen3.8 Flash Next’s context window?
Alibaba documents a context window of 262,144 tokens (256K). On AA-LCR, which tests reasoning across ~100,000-token document sets, it scores 79.7%.
Ratings: LMArena leaderboard dataset (CC BY 4.0), rescaled for the lens scores · leaderboards to 22 Sep 2026
Evaluations, prices and speed: Artificial Analysis (artificialanalysis.ai) · fetched 23 Sep 2026
Exchange rate: Reserve Bank of Australia, table F11.1 · A$1 = US$0.7123 on 22 Sep 2026 · prices exclude GST
Snapshot 23 Sep 2026 · updated weekly · How the rankings work →