AlibabaQwen3.5-4B
Qwen3.5-4B is an Alibaba open-weights reasoning model, released 24 Feb 2026. Its best configuration (thinking reasoning effort) ranks #481 of 785 on the BenchLeader Index at 46.6 ±5.2, in the lower half. The ± is the point: 335 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 long context (56) and lowest in knowledge (35). At $0.060 per million tokens blended it is among the cheapest fifth of ranked models. Output speed of 17 tokens per second puts it in the slowest quarter, with a first answer in 119.8 s. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 2 Mar 2026.stale: no new result in six months
- Blended price
- $0.060/M
- $0.030 in · $0.150 out
- Output speed
- 17 tok/s
- measured by Artificial Analysis
- First answer
- 120 s
- Context
- 66k
- Full answer
- 150 s
- median, reasoning included
- Cost per run
- –
- Released
- 24 Feb 2026
- Overall index47
- Reasoning41
- Agents & tools39
- Knowledge35
- Instruction following53
- Multimodal48
- Long context56
- Composite44
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 42.2 | #604 | 11 tok/s | 1.12 s | $0.0001 | Agents & tools 36 · Composite 42 · Instruction following 37 · Knowledge 28 · Long context 42 · Maths 48 · Multimodal 45 · Reasoning 41 |
| thinkingbest | 46.6 | #481 | 17 tok/s | 120 s | $0.0001 | Agents & tools 39 · Composite 44 · Instruction following 53 · Knowledge 35 · Long context 56 · Multimodal 48 · Reasoning 41 |
| not stated | – | – | – | – | $0.0004 | Maths 31 |
How its index has moved
9 Sept 2026 to 21 Oct 2026- thinking
- no reasoning
The 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
See all LLM benchmarks. 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 | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| GPQA Diamond (AA)not in index | 71.2%#288 | 77.1%#225 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 8.0%#355 | 9.9%#325 | Artificial Analysis | |
| CritPt | 0.0%#315 | 0.0%#315 | Artificial Analysis |
Coding
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Terminal-Bench 2.1 (AA)not in index | 21.4%#170 | 25.8%#167 | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Terminal-Bench Hard | 11.4%#215 | 18.2%#171 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 87.7%#76 | 92.1%#53 | Artificial Analysis | |
| τ³-Banking (AA) | 4.3%#193 | 6.8%#165 | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| OTIS Mock AIME | 55.8%#173 | – | Epoch AI Benchmarking Hub | |
| AIME 2026effort not statedpossible contamination | – | 89.7%#28 | MathArena | |
| HMMT February 2026effort not statedpossible contamination | – | 72.9%#28 | MathArena |
Knowledge
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-Omniscience | -74.7#527 | -58.4#453 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 11.2%#483 | 15.1%#420 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 3.2%#543 | 13.4%#362 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| IFBench | 33.3%#330 | 52.0%#164 | Artificial Analysis |
Multimodal
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| MMMU-Pro | 62.1%#199 | 65.4%#175 | Artificial Analysis |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 34.7%#377 | 63.0%#268 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 10.8#372 | 13.1#316 | 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 at $0.066 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0001 | $0.0001 | 2.3 min |
| Summarize a 30-page report | 12,000 / 600 | $0.0005 | $0.0008 | 2.6 min |
| Code edit | 6,000 / 1,500 | $0.0004 | $0.0006 | 3.5 min |
| Agentic coding session | 60,000 / 4,000 | $0.0024 | $0.0040 | 6.0 min |
| Structured extraction | 2,000 / 200 | $0.0001 | $0.0001 | 2.2 min |
More from Alibaba
See also
Questions people ask
- What are Qwen3.5-4B's benchmark scores?
- As of 24 Oct 2026, Qwen3.5-4B (thinking effort) scores 46.6 ±5.2 on the BenchLeader Index, strongest in long context (56) and weakest in knowledge (35). That rests on 8 independent benchmark results, and every score on this page links to its publisher.
- How does Qwen3.5-4B rank?
- Qwen3.5-4B (thinking effort) is #481 of 785 ranked configurations on 24 Oct 2026. 335 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Qwen3.5-4B cost per 1M tokens?
- As of 24 Oct 2026, Qwen3.5-4B costs $0.030 per 1M input tokens and $0.150 per 1M output tokens, $0.060 blended (three input tokens per output token).
- When was Qwen3.5-4B released?
- Alibaba released Qwen3.5-4B on 24 Feb 2026. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Qwen3.5-4B: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/qwen3-5-4b.