50 models7 providersprices scraped nightly — no estimates
alibaba · qwen familyVerified May 18, 2026

Qwen 3 72B Instruct

72B params · 131K context · qwen · released 2025-04 · 5 providers

Alibaba's Qwen 3 72B Instruct is the most credible alternative to Llama 3.3 70B in the same parameter class. Released in 2025, it ships with stronger multilingual support (particularly for Chinese, Japanese, Korean, and Arabic) and slightly better performance on long-context retrieval tasks. Pricing on hosted providers is broadly similar to Llama 3.3 70B — typically $0.25–$0.40 per 1M input tokens, $0.50–$0.90 per 1M output. Worth choosing over Llama for non-English-heavy workloads, code generation in non-English comments, or multilingual chat applications. For English-only deployments the choice between the two is largely a coin flip; pick whichever has more favorable pricing on your preferred provider.

Cheapest right now
DeepInfra$0.45/1M out
Go to DeepInfra ↗Compare
sticky while scrolling · verified May 18, 2026
Cheapest now
$0.45
DeepInfra · fp16 · $/1M out
90-day price Δ
—
cheapest-host trend
Fastest TTFT
400 ms
Fireworks AI · 230 tok/s
Providers
5
indexed providers
✓ verified May 18, 2026
#Provider$/1M in$/1M outTTFT p50tok/sQuant
1DeepInfraCHEAPEST$0.23$0.45435 ms195fp16go ↗
2Fireworks AIFASTEST$0.22$0.88400 ms230fp16go ↗
3Together AI$0.29$0.29——unknowngo ↗
4OpenRouter$0.27$0.85——unknowngo ↗
5Hyperbolic$0.40$0.40520 ms100fp16go ↗
spread $0.29 → $0.88Cost for your workload →CSV ↓

Similar models

same class · by param size or family
ModelParamsFrom $/1M outFamily
Qwen 2.5 72B Instruct72B—qwencompare →
Refact Llama 3.1 70B70B—refactcompare →
Qwen 3 8B Instruct8B—qwencompare →
Yi 1.5 9B Chat9B—yiview →

Run it locally instead?

fp16~158.4 GB VRAM
unknown~158.4 GB VRAM
Which rigs run this →

Hosted at $0.45/1M on DeepInfra. For most workloads, hosted is cheaper than local GPU power.

Questions developers ask

How much does it cost to run Qwen 3 72B Instruct for 100M tokens?▾

Running Qwen 3 72B Instruct with 100M input and 10M output tokens per month costs approximately $27.50 on DeepInfra, the cheapest available provider as of the latest pricing data. Costs vary significantly depending on your input/output ratio and whether you use prompt caching.

What is the cheapest provider for Qwen 3 72B Instruct?▾

DeepInfra currently offers Qwen 3 72B Instruct at the lowest total cost for a standard workload. Prices change frequently — check the provider table above for the latest data.

What context window does Qwen 3 72B Instruct support?▾

Qwen 3 72B Instruct supports a context window of 131,072 tokens. Individual providers may cap this lower.

What's the cheapest way to run Qwen 3 72B Instruct?▾

The cheapest way to run Qwen 3 72B Instruct is via DeepInfra, starting at $0.23 per million input tokens. If your workload is prompt-heavy, enabling prompt caching can reduce costs further.

Is there a free tier for Qwen 3 72B Instruct?▾

Free tiers vary by provider and change frequently. Check each provider's current pricing page for trial credits or free-tier limits. The prices shown on this page reflect paid API access.

How much does Qwen 3 72B Instruct cost for 1M tokens?▾

Qwen 3 72B Instruct input pricing starts at $0.23 per million tokens on DeepInfra. Output tokens are typically priced 2–4× higher than input tokens depending on the provider.

Keep exploring

Cheapest on Fireworks AI →Compare vs Qwen 2.5 72B Instruct →Compare vs Qwen 2.5 Coder 32B Instruct →Compare vs Qwen 2.5 Coder 7B Instruct →See full ranking: Cheapest input →See full ranking: Fastest TTFT →Used in the Customer support chatbot workload →Price history →
Methodology: prices scraped nightly from public pricing pages; snapshots append-only. Read methodology · Public API · CSV export · OG image per model.

Prices verified May 18, 2026 by scraping 5 providers. Methodology · Raw data