Qwen/Qwen2.5-72B-Instruct
alibaba/qwen2-5-72b-instructPor Alibaba (Qwen) · familia: qwen · lanzado 2024-09-18
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Capacidades
Model fit scores
0–100 · higher is betterThese scores reward declared capabilities, context size, price and provider availability — they are not benchmark results. Use them as a directional signal alongside your own evaluation.
Coding63
- Tool calling40/40
- Structured output20/20
- Reasoning0/10
- Context window (100K → 1M)0/20
- Provider availability3/10
Agents63
- Tool calling35/35
- Structured output25/25
- Reasoning0/15
- Output token limit0/15
- Provider availability3/10
JSON / structured output99
- Structured output / JSON mode50/50
- Tool calling20/20
- Temperature control10/10
- Price-friendly for high-volume19/20
Cost efficiency70
- Headline price (log-scaled)70/95
- Has prompt-cache pricing0/5
Long context0
- Context ≥ 100K0/100
Production-readiness73
- Number of independent providers15/40
- Has published per-token price20/20
- Context window ≥ 8K15/15
- No data inconsistencies across providers8/10
- Official model (not derivative)15/15
Cost Efficiency Index
Open full calculator →Estimated cost using the recommended provider's headline rate. Each scenario fixes average input/output tokens — the assumptions are shown in the third column.
| Scenario | Cost | Assumption |
|---|---|---|
RAG answer per 1,000 RAG answers | $0.74 < $0.01 per request | 5K input tokens (query + 4 retrieved chunks of ~1K each) and a 500-token answer. Typical SaaS knowledge-base bot. |
Support ticket triage per 10,000 tickets | $1.48 < $0.01 per request | 1K input tokens (ticket body + system prompt) and a 100-token JSON classification reply. High-volume customer support. |
Data extraction per 1,000 documents | $0.41 < $0.01 per request | 2K input tokens (a single document page) and a 500-token JSON extraction. ETL / invoice / form pipelines. |
Code review per 1,000 PRs | $1.26 < $0.01 per request | 8K input tokens (diff + surrounding files) and a 1K-token review comment. PR-bot workloads. |
Agent step per 1,000 steps | $1.55 < $0.01 per request | 12K input tokens (long-running tool history) and a 600-token tool-call decision. Cost per agent step. |
Detalle de precios
Precio recomendado de abacus · Qwen/Qwen2.5-72B-Instruct
Disponible en 3 proveedores
| Proveedor | ID de modelo del proveedor | Entrada / 1M | Salida / 1M | Contexto | Lanzado |
|---|---|---|---|---|---|
| SiliconFlow (China) siliconflow-cn | Qwen/Qwen2.5-72B-Instruct | $0.617 | $0.617 | 33K | 2024-09-18 |
| SiliconFlow siliconflow | Qwen/Qwen2.5-72B-Instruct | $0.590 | $0.590 | 33K | 2024-09-18 |
| Abacus abacus | Qwen/Qwen2.5-72B-Instruct | $0.110 | $0.380 | 128K | 2024-09-19 |
Inconsistencias de datos entre proveedores
- context_window varies: 128000, 33000
Los proveedores reportan valores distintos para este modelo. Los datos clave de arriba usan un proveedor representativo; consulta la tabla para detalles por proveedor.
Frequently asked questions
How much does Qwen/Qwen2.5-72B-Instruct cost?
Qwen/Qwen2.5-72B-Instruct costs $0.110 per 1M input tokens and $0.380 per 1M output tokens, sourced from abacus. Cache reads, audio tokens and >200K-context tiers (where applicable) are listed in the Pricing detail block above.
What is the context window of Qwen/Qwen2.5-72B-Instruct?
Qwen/Qwen2.5-72B-Instruct has a context window of 33K tokens, with a max output of 4K tokens per reply. This is the total combined size of prompt + completion.
Does Qwen/Qwen2.5-72B-Instruct support tool calling?
Yes. Qwen/Qwen2.5-72B-Instruct supports tool calling (function calling). This makes it suitable for production agent and automation workloads where the model has to invoke external functions reliably.
Does Qwen/Qwen2.5-72B-Instruct support structured output / JSON mode?
Yes. Qwen/Qwen2.5-72B-Instruct supports structured output / JSON-schema-constrained decoding. This makes it suitable for production agent and automation workloads where the model has to invoke external functions reliably.
Can Qwen/Qwen2.5-72B-Instruct accept image input?
No. Qwen/Qwen2.5-72B-Instruct only accepts text as input. If you need image input, see our /capabilities/vision list for current vision-capable models.
Is Qwen/Qwen2.5-72B-Instruct open-weight?
Yes. Qwen/Qwen2.5-72B-Instruct's weights are publicly available, so you can self-host or fine-tune. Note that open weights ≠ open source — the training data and code are typically not released.
What are the best alternatives to Qwen/Qwen2.5-72B-Instruct?
If Qwen/Qwen2.5-72B-Instruct doesn't fit, consider Qwen3.8 27B, Qwen3.8 Max, Qwen3.6 35B-A3B. Each one targets the same use case — see the Related links below for direct head-to-head pages.
Where does this data come from?
All numbers are normalised into a single canonical model record and reconciled with each provider's official documentation. We re-pull daily and write any changes (price, context, capability) to the /changelog page.
Explore more
More Alibaba (Qwen) models
- Qwen3.8 27B$0.00 in / $0.00 out
- Qwen3.8 Max$2.00 in / $6.00 out
- Qwen3.6 35B-A3B$0.38 in / $2.25 out
- Qwen3.8 Flash$0.15 in / $0.47 out
- Qwen3.7 Max$2.50 in / $7.50 out
Capability lists this model is in
Última actualización:
Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.