Qwen: Qwen3 14B
Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode …
qwen/qwen3-14b-04-28
Qwen3-14B is a dense 14.8B parameter causal language model from the Qwen3 series, designed for both complex reasoning and efficient dialogue. It supports seamless switching between a "thinking" mode for tasks like math, programming, and logical inference, and a "non-thinking" mode for general-purpose conversation. The model is fine-tuned for instruction-following, agent tool use, creative writing, and multilingual tasks across 100+ languages and dialects. It natively handles 32K token contexts and can extend to 131K tokens using YaRN-based scaling.
Specifications & capabilities
- Input
- text
- Output
- text
- Context window · tokens
- 40960
- Published model ID
- Qwen3-14B
- Capabilities
- reasoning, structuredOutputs, toolCalling
Description source ↗Evaluations
1–20 / 25
89.46%
- source id
- bfcl
- leaderboard name
- BFCL Non-Live AST
- dimension
- toolUse
- metric
- Non-live AST accuracy
- rank
- 7
Original result ↗59.788%
- source id
- openRouterBenchmarks
- leaderboard name
- GPQA Diamond · Accuracy · NextBit
- dimension
- knowledge
- metric
- GPQA Diamond · Accuracy
- rank
- 9
- sample size
- 1386
- aggregation scope
- status
- published
- harness
- url
- https://github.com/OpenRouterTeam/benchmark-harness
- name
- OpenRouter benchmark harness
- direction
- higherIsBetter
- run_count
- 7
- conditions
- window
- newest 90 days; successful visible nonzero runs
- dataset
- benchmark-leaderboard
- benchmark
- id
- gpqa_diamond
- name
- GPQA Diamond
- record_kind
- aggregate
- provider_name
- NextBit
- attribution
- Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
- last_run_at
- 2026-11-01T11:08:51.026Z
- license_url
- https://creativecommons.org/licenses/by/4.0/
- methodology
- Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
- metric_kind
- GPQA Diamond · Accuracy
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- sample_size
- 1386
- source_file
- benchmark-leaderboard.json
- source_page
- https://openrouter.ai/benchmarks/gpqa-diamond
- measurements
- accuracy
- 0.5978837142857143
- accuracy_std_dev
- 0.01426892240114143
- cost_per_task_usd
- 0.0030797525252525253
- output_tokens_per_task
- 13628.22077922078
- latency_seconds_per_task
- 643.4021818181818
- published_at
- 2026-11-16T01:46:02.653Z
- snapshot_date
- 2026-11-16
- source_revision
- 7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
- transformations
- TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
- metric_population
- 10
- benchmark_evaluation
- true
- openrouter_permaslug
- qwen/qwen3-14b-04-28
- source model id
- qwen/qwen3-14b-04-28
Original result ↗80.01%
- source id
- bfcl
- leaderboard name
- BFCL Live
- dimension
- toolUse
- metric
- Live accuracy
- rank
- 13
Original result ↗79.35%
- source id
- bfcl
- leaderboard name
- BFCL Live
- dimension
- toolUse
- metric
- Live accuracy
- rank
- 16
Original result ↗87.18%
- source id
- bfcl
- leaderboard name
- BFCL Irrelevance Detection
- dimension
- toolUse
- metric
- Irrelevance detection
- rank
- 22
Original result ↗49.242%
- source id
- openRouterBenchmarks
- leaderboard name
- GPQA Diamond · Accuracy · Alibaba
- dimension
- knowledge
- metric
- GPQA Diamond · Accuracy
- rank
- 29
- sample size
- 396
- aggregation scope
- status
- published
- harness
- url
- https://github.com/OpenRouterTeam/benchmark-harness
- name
- OpenRouter benchmark harness
- direction
- higherIsBetter
- run_count
- 2
- conditions
- window
- newest 90 days; successful visible nonzero runs
- dataset
- benchmark-leaderboard
- benchmark
- id
- gpqa_diamond
- name
- GPQA Diamond
- record_kind
- aggregate
- provider_name
- Alibaba
- attribution
- Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
- last_run_at
- 2026-10-01T16:05:56.758Z
- license_url
- https://creativecommons.org/licenses/by/4.0/
- methodology
- Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
- metric_kind
- GPQA Diamond · Accuracy
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- sample_size
- 396
- source_file
- benchmark-leaderboard.json
- source_page
- https://openrouter.ai/benchmarks/gpqa-diamond
- measurements
- accuracy
- 0.49242400000000003
- accuracy_std_dev
- 0.0008419999999999817
- cost_per_task_usd
- 0.0017634166666666666
- output_tokens_per_task
- 1731.7373737373737
- latency_seconds_per_task
- 30.89138888888889
- published_at
- 2026-11-16T01:46:02.653Z
- snapshot_date
- 2026-11-16
- source_revision
- 7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
- transformations
- TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
- metric_population
- 29
- benchmark_evaluation
- true
- openrouter_permaslug
- qwen/qwen3-14b-04-28
- source model id
- qwen/qwen3-14b-04-28
Original result ↗87.5%
- source id
- bfcl
- leaderboard name
- BFCL Relevance Detection
- dimension
- toolUse
- metric
- Relevance detection
- rank
- 34
Original result ↗34.75%
- source id
- bfcl
- leaderboard name
- BFCL Multi Turn
- dimension
- agentic
- metric
- Multi-turn accuracy
- rank
- 35
Original result ↗41.03%
- source id
- bfcl
- leaderboard name
- BFCL Overall
- dimension
- toolUse
- metric
- Overall accuracy
- rank
- 43
Original result ↗26.13%
- source id
- bfcl
- leaderboard name
- BFCL Multi Turn
- dimension
- agentic
- metric
- Multi-turn accuracy
- rank
- 45
Original result ↗19.57%
- source id
- bfcl
- leaderboard name
- BFCL Memory
- dimension
- agentic
- metric
- Memory accuracy
- rank
- 45
Original result ↗84.94%
- source id
- bfcl
- leaderboard name
- BFCL Non-Live AST
- dimension
- toolUse
- metric
- Non-live AST accuracy
- rank
- 46
Original result ↗37.77%
- source id
- bfcl
- leaderboard name
- BFCL Overall
- dimension
- toolUse
- metric
- Overall accuracy
- rank
- 47
Original result ↗81.94%
- source id
- bfcl
- leaderboard name
- BFCL Irrelevance Detection
- dimension
- toolUse
- metric
- Irrelevance detection
- rank
- 49
Original result ↗61.234%
- source id
- openRouterBenchmarks
- leaderboard name
- GPQA Diamond · Accuracy · DeepInfra
- dimension
- knowledge
- metric
- GPQA Diamond · Accuracy
- rank
- 55
- sample size
- 589
- aggregation scope
- status
- published
- harness
- url
- https://github.com/OpenRouterTeam/benchmark-harness
- name
- OpenRouter benchmark harness
- direction
- higherIsBetter
- run_count
- 3
- conditions
- window
- newest 90 days; successful visible nonzero runs
- dataset
- benchmark-leaderboard
- benchmark
- id
- gpqa_diamond
- name
- GPQA Diamond
- record_kind
- aggregate
- provider_name
- DeepInfra
- attribution
- Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
- last_run_at
- 2026-10-08T22:16:42.056Z
- license_url
- https://creativecommons.org/licenses/by/4.0/
- methodology
- Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
- metric_kind
- GPQA Diamond · Accuracy
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- sample_size
- 589
- source_file
- benchmark-leaderboard.json
- source_page
- https://openrouter.ai/benchmarks/gpqa-diamond
- measurements
- accuracy
- 0.6123375670628184
- accuracy_std_dev
- 0.008273645159312946
- cost_per_task_usd
- 0.003321100169779287
- output_tokens_per_task
- 13430.818336162989
- latency_seconds_per_task
- 426.79844821731746
- published_at
- 2026-11-16T01:46:02.653Z
- snapshot_date
- 2026-11-16
- source_revision
- 7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
- transformations
- TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
- metric_population
- 71
- benchmark_evaluation
- true
- openrouter_permaslug
- qwen/qwen3-14b-04-28
- source model id
- qwen/qwen3-14b-04-28
Original result ↗10.5%
- source id
- bfcl
- leaderboard name
- BFCL Web Search
- dimension
- agentic
- metric
- Web-search accuracy
- rank
- 55
Original result ↗42.556%
- source id
- openRouterBenchmarks
- leaderboard name
- tau2-bench Verified Airline · Accuracy · DeepInfra
- dimension
- agentic
- metric
- tau2-bench Verified Airline · Accuracy
- rank
- 56
- sample size
- 150
- aggregation scope
- status
- published
- harness
- url
- https://github.com/OpenRouterTeam/benchmark-harness
- name
- OpenRouter benchmark harness
- direction
- higherIsBetter
- run_count
- 3
- conditions
- window
- newest 90 days; successful visible nonzero runs
- dataset
- benchmark-leaderboard
- benchmark
- id
- tau_bench_verified_airline
- name
- tau2-bench Verified Airline
- record_kind
- aggregate
- provider_name
- DeepInfra
- attribution
- Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
- last_run_at
- 2026-10-08T22:41:10.664Z
- license_url
- https://creativecommons.org/licenses/by/4.0/
- methodology
- Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
- metric_kind
- tau2-bench Verified Airline · Accuracy
- rank_method
- competition rank among reported evaluations in identical conditions, including unlinked observations
- sample_size
- 150
- source_file
- benchmark-leaderboard.json
- source_page
- https://openrouter.ai/benchmarks/tau2-bench-airline
- measurements
- accuracy
- 0.4255553333333334
- accuracy_std_dev
- 0.034462310778910675
- cost_per_task_usd
- 0.029985173333333334
- output_tokens_per_task
- 18074.12
- latency_seconds_per_task
- 822.67542
- published_at
- 2026-11-16T01:46:02.653Z
- snapshot_date
- 2026-11-16
- source_revision
- 7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
- transformations
- TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
- metric_population
- 63
- benchmark_evaluation
- true
- openrouter_permaslug
- qwen/qwen3-14b-04-28
- source model id
- qwen/qwen3-14b-04-28
Original result ↗81.25%
- source id
- bfcl
- leaderboard name
- BFCL Relevance Detection
- dimension
- toolUse
- metric
- Relevance detection
- rank
- 57
Original result ↗10%
- source id
- bfcl
- leaderboard name
- BFCL Web Search
- dimension
- agentic
- metric
- Web-search accuracy
- rank
- 57
Original result ↗11.18%
- source id
- bfcl
- leaderboard name
- BFCL Memory
- dimension
- agentic
- metric
- Memory accuracy
- rank
- 70
Original result ↗Published prices
1–10 / 10
$0.07
- Inference provider
- cheaperInference
- Platform
- Cheaper Inference
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- zdr
- false
- protocol
- chat_completions
- service_tier
- standard
- Billing label
- Input tokens (per 1M)
$0.17
- Inference provider
- cheaperInference
- Platform
- Cheaper Inference
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- zdr
- false
- protocol
- chat_completions
- service_tier
- standard
- Billing label
- Output tokens (per 1M)
$0.10
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- input_tokens_per_million
- Currency
- USD
$0.22
- Inference provider
- openRouter
- Platform
- OpenRouter
- Unit
- output_tokens_per_million
- Currency
- USD
$0.07
- Inference provider
- siliconFlow
- Platform
- SiliconFlow
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing label
- Input tokens (per 1M)
$0.28
- Inference provider
- siliconFlow
- Platform
- SiliconFlow
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing label
- Output tokens (per 1M)
$0.12
- Inference provider
- vercelAIGatewayCatalog
- Platform
- deepinfra
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- endpoint
- deepinfra | alibaba/qwen-3-14b
- model_type
- language
- quantization
- fp8
- billing_scope
- via Vercel AI Gateway
- Billing label
- Input tokens / 1M
- Pricing via
- Vercel AI Gateway
$0.24
- Inference provider
- vercelAIGatewayCatalog
- Platform
- deepinfra
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- endpoint
- deepinfra | alibaba/qwen-3-14b
- model_type
- language
- quantization
- fp8
- billing_scope
- via Vercel AI Gateway
- Billing label
- Output tokens / 1M
- Pricing via
- Vercel AI Gateway
$0.12
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- input_tokens_per_million
- Currency
- USD
- Billing conditions
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Input tokens / 1M
- Pricing via
- Vercel AI Gateway
$0.24
- Inference provider
- vercelAIGatewayCatalog
- Platform
- Vercel AI Gateway
- Unit
- output_tokens_per_million
- Currency
- USD
- Billing conditions
- model_type
- language
- quote_scope
- gateway_catalog
- billing_scope
- Vercel public catalog; endpoint prices may differ
- Billing label
- Output tokens / 1M
- Pricing via
- Vercel AI Gateway
Provider specifications & access
1–11 / 11
- source model id
- alibaba/qwen-3-14b
- hosting
- vercel-gateway
- context length
- 40960
- max output tokens
- 16384
- capabilities
- reasoning
- structuredOutputs
- toolCalling
- source model id
- azure-huggingface/qwen--qwen3-14b@1
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- azure-huggingface/qwen--qwen3-14b@2
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- azure-huggingface/qwen--qwen3-14b@3
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- azure-huggingface/qwen--qwen3-14b@4
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- azure-huggingface/qwen--qwen3-14b@5
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- azure-huggingface/qwen--qwen3-14b@6
- hosting
- foundry-managed-compute
- license
- apache-2.0
- context length
- 40960
- source model id
- publishers/hf-qwen/models/qwen3-14b@001
- hosting
- vertex-self-deployed
- source model id
- publishers/qwen/models/qwen3@qwen3-14b
- hosting
- vertex-self-deployed
- source model id
- qwen-3-14b
- hosting
- gateway
- context length
- 40960
- max output tokens
- 16384
- capabilities
- reasoning
- structuredOutputs
- toolCalling
- source model id
- Qwen/Qwen3-14B
- hosting
- siliconflow-hosted
- max output tokens
- 131072
- capabilities
- reasoning
- structuredOutputs
- toolCalling
Safety assessments
1–1 / 1
26 / 100
- source id
- stanfordFMTI
- canonical model id
- qwen/qwen3-14b-04-28
- taxonomy
- transparency
- hazard category
- transparency
- hazard name
- Organization transparency
- policy area
- Alibaba
- assessment type
- organization-level disclosure
- metric
- disclosure_score
- result label
- 26 / 100
- sample size
- 100
- updated at
- 2026-11-09T18:08:05.227Z
Source specification observations
1–14 / 14
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- azureFoundry
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- context_length
- value
- 131K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
- field
- context_length
- value
- 40960
- normalized value
- 40960
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
- field
- knowledge_cutoff
- reported field
- knowledge_cutoff
- value
- 2025-03-31
- normalized value
- 2025-03-31
- precision
- day
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- knowledge_cutoff
- reported field
- knowledge_cutoff
- value
- 2025-04
- normalized value
- 2025-04
- precision
- month
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
- field
- max_output_tokens
- value
- 36864
- normalized value
- 36864
- scope
- source-model
- source id
- openRouter
- confidence
- 1
- field
- max_output_tokens
- value
- 131K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
- field
- max_output_tokens
- value
- 16384
- normalized value
- 16384
- scope
- source-model
- source id
- vercelAIGatewayCatalog
- confidence
- 1
Hugging Face model card
1–1 / 1
- downloads
- 12939734
- likes
- 538
- pipeline tag
- text-generation
- library name
- transformers
- License
- apache-2.0
Explore more models