Qwen3 30B A3B vs Qwen3-14B API pricing
Updated
Qwen3 30B A3B: USD 0.051 input, USD 0.335 output per 1M tokens (Cloudflare Workers AI). Qwen3-14B: USD 0.07 input, USD 0.28 output per 1M tokens (SiliconFlow). At 3 input tokens per output token, Qwen3-14B costs 0.4% more than Qwen3 30B A3B (USD 0.1225 vs USD 0.122 per 1M).
How do Qwen3 30B A3B and Qwen3-14B differ on price?
- Cheapest input price: Qwen3 30B A3B USD 0.051 against USD 0.07 per 1M tokens, so Qwen3-14B costs 1.37× as much (37% more).
- Cheapest output price: Qwen3-14B USD 0.24 against USD 0.25 per 1M tokens, so Qwen3 30B A3B costs 1.04× as much (4% more).
- Among the 201 indexed models, Qwen3 30B A3B ranks 28th cheapest and Qwen3-14B 29th on the 3:1 blend.
- The cheaper model changes at 0.25 input tokens per output token.
- Neither model has a standard price with a cached-input discount, so prompt caching does not lower either bill.
- Qwen3-14B states up to 131,072 tokens of context, 3.2× the 40,960 of Qwen3 30B A3B.
- Both have 3 providers with a standard price; 1 provider sells both (DeepInfra).
- On one account, DeepInfra charges USD 0.215 for Qwen3 30B A3B (1.76× its cheapest price) and USD 0.15 for Qwen3-14B (1.22× its cheapest price) at 3:1, the lowest combined price among the providers that sell both.
- Shopping around matters more for Qwen3 30B A3B: its dearest standard price is 1.76× its cheapest, against 1.22× for Qwen3-14B.
- For the one-developer coding agent preset, the gap is USD 351.00 a year in Qwen3 30B A3B's favour (Qwen3 30B A3B USD 121.95 a month, Qwen3-14B USD 151.20).
- Per character of input, Qwen3 30B A3B costs 27% less than Qwen3-14B in every measured language, because both use the Qwen3 32B tokenizer.
Qwen3 30B A3B and Qwen3-14B prices side by side
USD per 1M tokens, standard tier. "Cheapest" is the lowest price at 3 input tokens to 1 output token; each line can come from a different provider.
| Price | Qwen3 30B A3B | Qwen3-14B |
|---|---|---|
| Model page | Qwen3 30B A3B (fp8) Qwen · 3 providers | Qwen3-14B Qwen · 3 providers |
| Cheapest standard price, blended 3:1 | USD 0.122 lower USD 0.051 in / USD 0.335 out at Cloudflare Workers AI Cloudflare Workers AI page checked 2026-11-14 | USD 0.1225 USD 0.07 in / USD 0.28 out at SiliconFlow SiliconFlow page checked 2026-11-14 |
| Cheapest input | USD 0.051 lower Cloudflare Workers AI source | USD 0.07 SiliconFlow source |
| Cheapest cached input | No standard price lists a cache price; cached tokens bill at the input price. | No standard price lists a cache price; cached tokens bill at the input price. |
| Cheapest output | USD 0.25 GMI Cloud source | USD 0.24 lower DeepInfra source |
| Maker's own API | Qwen publishes no paid standard price on its own API that we found. | Qwen publishes no paid standard price on its own API that we found. |
| Median standard price | USD 0.08 / USD 0.335 input / output, 3 prices | USD 0.12 / USD 0.24 input / output, 3 prices |
| Context window | 40,960 tokens, largest stated; from 40,946 at some providers | 131,072 tokens, largest stated; from 32,768 at some providers |
| Providers with a standard price | 3 | 3 |
- Lower marks the lower price on a line; tie, equal prices. Other tiers and free rows are on the model pages.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What do Qwen3 30B A3B and Qwen3-14B cost for a real workload?
30 days at each model's cheapest standard price for the workload (presets of the cost calculator; cached input at the cached price where listed).
| Workload | Qwen3 30B A3B | Qwen3-14B | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 10.62 Cloudflare Workers AI | USD 11.34 SiliconFlow | USD 0.72 Qwen3 30B A3B lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 17.26 Cloudflare Workers AI | USD 21.00 SiliconFlow | USD 3.74 Qwen3 30B A3B lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 121.95 Cloudflare Workers AI | USD 151.20 SiliconFlow | USD 29.25 Qwen3 30B A3B lower |
Qwen3 30B A3B is cheaper on all three workloads. The widest gap is on Coding agent (one developer): the dearer model costs 24% more than the cheaper one.
Does the cheaper model change with your input:output ratio?
The cheaper model changes with the mix: below 0.25 input tokens per output token, Qwen3-14B is cheaper; above 0.25, Qwen3 30B A3B is cheaper. At the site's 3:1 blend, Qwen3 30B A3B is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
| Input : output tokens | Qwen3 30B A3B | Qwen3-14B |
|---|---|---|
| 1 : 3 (long answers) | USD 0.2075 lower GMI Cloud | USD 0.21 DeepInfra |
| 1 : 1 | USD 0.165 lower GMI Cloud | USD 0.175 SiliconFlow |
| 3 : 1 (the site’s blend) | USD 0.122 lower Cloudflare Workers AI | USD 0.1225 SiliconFlow |
| 10 : 1 (retrieval, long prompts) | USD 0.07682 lower Cloudflare Workers AI | USD 0.08909 SiliconFlow |
| 30 : 1 (agents, big contexts) | USD 0.06016 lower Cloudflare Workers AI | USD 0.07677 SiliconFlow |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do Qwen3 30B A3B and Qwen3-14B cost per character in other languages?
Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the Qwen3 32B tokenizer for Qwen3 30B A3B (the same one for both, so the gap equals the price gap):
| Content | Qwen3 30B A3B | Qwen3-14B |
|---|---|---|
| English | USD 0.011 lower Cloudflare Workers AI | USD 0.015 SiliconFlow |
| Arabic MSA | USD 0.020 lower Cloudflare Workers AI | USD 0.027 SiliconFlow |
| Egyptian Arabic | USD 0.020 lower Cloudflare Workers AI | USD 0.028 SiliconFlow |
| Hindi | USD 0.047 lower Cloudflare Workers AI | USD 0.065 SiliconFlow |
| Chinese | USD 0.033 lower Cloudflare Workers AI | USD 0.045 SiliconFlow |
| Spanish | USD 0.014 lower Cloudflare Workers AI | USD 0.019 SiliconFlow |
| Code | USD 0.013 lower Cloudflare Workers AI | USD 0.018 SiliconFlow |
From the real cost by language dataset (measured tokens per character on FLORES-200 and other texts). Your own text will differ.
Check it on your own traffic
- Qwen3 30B A3B (fp8) API pricing: all 3 providers and other tiers.
- Qwen3-14B API pricing: all 3 providers and other tiers.
- Cost calculator with Qwen3 30B A3B and Qwen3-14B preloaded
- Qwen3 token counter
- Prompt caching calculator
More comparisons: Gemma 4 26B A4B vs Qwen3 30B A3B · Qwen3.5 35B A3B vs Qwen3 30B A3B · Qwen3-30B-A3B-2507 vs Qwen3 30B A3B · Qwen3 32B vs Qwen3 30B A3B · Gemma 4 26B A4B vs Qwen3-14B · Qwen3-14B vs Qwen3-8B · every Qwen3 30B A3B comparison · every Qwen3-14B comparison · all 662 price comparisons
Questions
Is Qwen3 30B A3B cheaper than Qwen3-14B?
Yes. On a blend of 3 input tokens to 1 output token, Qwen3 30B A3B costs USD 0.122 per 1M tokens and Qwen3-14B USD 0.1225, at each model's cheapest standard price. For input alone, Qwen3 30B A3B is cheaper (USD 0.051 against USD 0.07 per 1M); for output alone, Qwen3-14B is (USD 0.24 against USD 0.25). This compares price only, not quality.
What do Qwen3 30B A3B and Qwen3-14B cost per 1M tokens?
Qwen3 30B A3B: from USD 0.051 input and USD 0.335 output at Cloudflare Workers AI; median of 3 standard prices USD 0.08 / USD 0.335. Qwen3-14B: from USD 0.07 input and USD 0.28 output at SiliconFlow; median of 3 standard prices USD 0.12 / USD 0.24. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both Qwen3 30B A3B and Qwen3-14B?
1 provider has a standard price for both: DeepInfra (USD 0.215 and USD 0.15 at 3:1). Buying each model from its own cheapest provider costs USD 0.122 and USD 0.1225.
Which is cheaper for a coding agent or a chatbot, Qwen3 30B A3B or Qwen3-14B?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 121.95 a month on Qwen3 30B A3B (Cloudflare Workers AI) against USD 151.20 on Qwen3-14B (SiliconFlow), a difference of USD 29.25. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 10.62 a month on Qwen3 30B A3B (Cloudflare Workers AI) against USD 11.34 on Qwen3-14B (SiliconFlow), a difference of USD 0.72. Each uses the cheapest standard price for that workload, cached input included. Try your own numbers.
At what input:output ratio does the cheaper of Qwen3 30B A3B and Qwen3-14B change?
The cheaper model changes with the mix: below 0.25 input tokens per output token, Qwen3-14B is cheaper; above 0.25, Qwen3 30B A3B is cheaper. At the site's 3:1 blend, Qwen3 30B A3B is cheaper. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
Do Qwen3 30B A3B and Qwen3-14B have cached input pricing?
Qwen3 30B A3B: no standard price lists a cached-input discount. Qwen3-14B: no standard price lists a cached-input discount. Cached tokens are billed at the input price where no cache price is published.
Which has the larger context window, Qwen3 30B A3B or Qwen3-14B?
Qwen3-14B: up to 131,072 tokens, against 40,960 for Qwen3 30B A3B, as stated on providers' pricing pages. Some providers offer less than the maximum; check the model pages before you pick the cheapest row for long prompts.