AlibabaQwen3 14B
Qwen3 14B is an Alibaba open-weights model, released 29 Apr 2025. Its best configuration ranks #472 of 785 on the BenchLeader Index at 46.9 ±7.7, in the lower half. The ± is the point: 450 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 agents & tools (53) and lowest in coding (29). At $0.251 per million tokens blended it is cheaper than most ranked models. Output speed of 80 tokens per second puts it slower than most, with a first token in 0.7 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 29 Apr 2025.stale: no new result in six months
- Blended price
- $0.251/M
- $0.144 in · $0.574 out
- Output speed
- 80 tok/s
- OpenRouter traffic, 7-day median; not yet measured by Artificial Analysis
- First answer
- 0.71 s
- Context
- 131k
- Full answer
- –
- Cost per run
- –
- Released
- 29 Apr 2025
- Overall index47
- Reasoning47
- Coding29
- Agents & tools53
- Maths52
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 | 37.9 | #723 | 76 tok/s | 0.91 s | $0.0001 | Agents & tools 38 · Composite 37 · Instruction following 29 · Knowledge 31 · Long context 23 · Maths 35 · Reasoning 41 |
| thinking | 40.4 | #657 | 76 tok/s | 27 s | $0.0001 | Agents & tools 34 · Composite 37 · Instruction following 43 · Knowledge 39 · Long context 23 · Reasoning 41 |
| not statedbest | 46.9 | #472 | 80 tok/s | 0.71 s | $0.0002 | Agents & tools 53 · Coding 29 · Maths 52 · Reasoning 47 |
How its index has moved
9 Sept 2026 to 23 Oct 2026- not stated
- 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 | not stated | Source | Trend |
|---|---|---|---|---|---|
| GPQA Diamond | 53.4%#208 | – | 63.8%#187 | Epoch AI Benchmarking Hub | |
| GPQA Diamond (AA)not in index | 47.0%#444 | 60.4%#373 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 4.1%#540 | 4.5%#496 | – | Artificial Analysis | |
| Kagi LLM Benchmarkeffort not stated | – | – | 49.1%#86 | Kagi LLM Benchmark | |
| CritPt | 0.0%#315 | 0.0%#315 | – | Artificial Analysis | |
| LMCAnot in indexeffort not stated | – | – | 18.2%#162 | Epoch AI Benchmarking Hub | |
| DTBenchnot in indexeffort not stated | – | – | 64.0%#170 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | thinking | not stated | Source | Trend |
|---|---|---|---|---|
| SciCodeeffort not stated | – | 31.6%#197 | SciCode | |
| SciCode (AA)not in index | 30.7%#210 | – | Artificial Analysis | |
| Terminal-Bench 4.0 (AA)not in index | 0.0%#152 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 4.9%#206 | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bench Hard | 5.3%#277 | 3.8%#298 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 32.2%#248 | 34.5%#238 | – | Artificial Analysis | |
| BFCL Overalleffort not stated | – | – | 41.0%#35 | Berkeley Function Calling Leaderboard | |
| τ³-Banking (AA) | – | 5.6%#175 | – | Artificial Analysis | |
| AutomationBenchnot in index | – | 0.2%#219 | – | Artificial Analysis | |
| GDP.pdf (AA)not in index | – | 1.2%#198 | – | Artificial Analysis |
Maths
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| OTIS Mock AIME | 25.8%#214 | 66.4%#152 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-Omniscience | -66.7#505 | -49.0#381 | Artificial Analysis | |
| AA-Omniscience: accuracynot in index | 13.3%#457 | 15.3%#415 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 7.7%#477 | 24.2%#249 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| IFBench | 23.9%#397 | 40.5%#256 | Artificial Analysis |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 0.0%#493 | 0.0%#493 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | – | 138.2#122 | Epoch AI Benchmarking Hub | |
| AA Intelligence Index v4.3.2 | 6.7#510 | 6.4#533 | – | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| DeepInfra | 89 tok/s | 0.56 s | $0.120 | $0.240 | 41k | fp8 |
| NextBit | 62 tok/s | 1.22 s | $0.100 | $0.220 | 41k | int4 |
Price history
The price above, per 1M tokens, every day it changed.
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 | – | 4.5 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0021 | – | 8.2 s |
| Code edit | 6,000 / 1,500 | $0.0017 | – | 19.5 s |
| Agentic coding session | 60,000 / 4,000 | $0.011 | – | 50.7 s |
| Structured extraction | 2,000 / 200 | $0.0004 | – | 3.2 s |
More from Alibaba
See also
Questions people ask
- What are Qwen3 14B's benchmark scores?
- As of 25 Oct 2026, Qwen3 14B scores 46.9 ±7.7 on the BenchLeader Index, strongest in agents & tools (53) and weakest in coding (29). That rests on 5 independent benchmark results, and every score on this page links to its publisher.
- How does Qwen3 14B rank?
- Qwen3 14B is #472 of 785 ranked configurations on 25 Oct 2026. 450 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 14B cost per 1M tokens?
- As of 25 Oct 2026, Qwen3 14B costs $0.144 per 1M input tokens and $0.574 per 1M output tokens, $0.251 blended (three input tokens per output token). The list price last changed on 25 Oct 2026.
- When was Qwen3 14B released?
- Alibaba released Qwen3 14B on 29 Apr 2025. BenchLeader has results for it at 3 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Qwen3 14B: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/qwen3-14b.