GPT-5 Nano vs Qwen3-30B-A3B-2507 API pricing
Updated
GPT-5 Nano: USD 0.05 input, USD 0.40 output per 1M tokens (OpenAI and 2 others). Qwen3-30B-A3B-2507: USD 0.09 input, USD 0.30 output per 1M tokens (SiliconFlow). At 3 input tokens per output token, Qwen3-30B-A3B-2507 costs 3.6% more than GPT-5 Nano (USD 0.1425 vs USD 0.1375 per 1M).
How do GPT-5 Nano and Qwen3-30B-A3B-2507 differ on price?
- Cheapest input price: GPT-5 Nano USD 0.05 against USD 0.09 per 1M tokens, so Qwen3-30B-A3B-2507 costs 1.8× as much (80% more).
- Cheapest output price: Qwen3-30B-A3B-2507 USD 0.30 against USD 0.40 per 1M tokens, so GPT-5 Nano costs 1.33× as much (33% more).
- Among the 201 indexed models, GPT-5 Nano ranks 33rd cheapest and Qwen3-30B-A3B-2507 36th on the 3:1 blend.
- The cheaper model changes at 2.5 input tokens per output token.
- Only GPT-5 Nano lists a cached-input discount: GPT-5 Nano at 3 standard prices of 3, from USD 0.005 (OpenAI, 90% below that provider's input price). Qwen3-30B-A3B-2507's cached tokens bill at the input price.
- Only Qwen3-30B-A3B-2507 has a stated context window (262,144 tokens); no provider of GPT-5 Nano states one on its pricing page.
- GPT-5 Nano has 3 providers with a standard price, 1 more than Qwen3-30B-A3B-2507; no provider sells both, so using both means two accounts.
- Shopping around matters more for Qwen3-30B-A3B-2507: its dearest standard price is 1.45× its cheapest, against 1× for GPT-5 Nano.
- OpenAI's own API is one of 3 providers tied for the lowest standard price we found for GPT-5 Nano, USD 0.1375 at 3:1.
- GPT-5 Nano has a batch price at 1 provider, from USD 0.06875 at 3:1 (OpenAI); Qwen3-30B-A3B-2507 has no batch price on file.
- For the one-developer coding agent preset, the gap is USD 1,582 a year in GPT-5 Nano's favour (GPT-5 Nano USD 57.15 a month, Qwen3-30B-A3B-2507 USD 189.00).
- Per character of input, the gap depends on the language: GPT-5 Nano costs 33% less than Qwen3-30B-A3B-2507 on Chinese text but 81% less than Qwen3-30B-A3B-2507 on Hindi, because the two tokenizers split those languages differently.
GPT-5 Nano and Qwen3-30B-A3B-2507 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 | GPT-5 Nano | Qwen3-30B-A3B-2507 |
|---|---|---|
| Model page | GPT-5 Nano OpenAI · 3 providers | Qwen3-30B-A3B-Instruct-2507 Qwen · 2 providers |
| Cheapest standard price, blended 3:1 | USD 0.1375 lower USD 0.05 in / USD 0.40 out at OpenAI and 2 others OpenAI page checked 2026-11-14 | USD 0.1425 USD 0.09 in / USD 0.30 out at SiliconFlow SiliconFlow page checked 2026-11-14 |
| Cheapest input | USD 0.05 lower OpenAI source | USD 0.09 SiliconFlow source |
| Cheapest cached input | USD 0.005 OpenAI source | No standard price lists a cache price; cached tokens bill at the input price. |
| Cheapest output | USD 0.40 OpenAI source | USD 0.30 lower SiliconFlow source |
| Maker's own API | USD 0.05 / USD 0.40 OpenAI source | Qwen publishes no paid standard price on its own API that we found. |
| Median standard price | USD 0.05 / USD 0.40 input / output, 3 prices | USD 0.1075 / USD 0.375 input / output, 2 prices |
| Context window | No provider states one on its pricing page. | 262,144 tokens, largest stated |
| Providers with a standard price | 3 | 2 |
- 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 GPT-5 Nano and Qwen3-30B-A3B-2507 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 | GPT-5 Nano | Qwen3-30B-A3B-2507 | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 9.27 OpenAI | USD 13.50 SiliconFlow | USD 4.23 GPT-5 Nano lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 14.76 OpenAI | USD 26.10 SiliconFlow | USD 11.34 GPT-5 Nano lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 57.15 OpenAI | USD 189.00 SiliconFlow | USD 131.85 GPT-5 Nano lower |
GPT-5 Nano is cheaper on all three workloads. The widest gap is on Coding agent (one developer): the dearer model costs 3.3× as much as the cheaper one.
Does the cheaper model change with your input:output ratio?
The cheaper model changes with the mix: below 2.5 input tokens per output token, Qwen3-30B-A3B-2507 is cheaper; above 2.5, GPT-5 Nano is cheaper. At the site's 3:1 blend, GPT-5 Nano 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 | GPT-5 Nano | Qwen3-30B-A3B-2507 |
|---|---|---|
| 1 : 3 (long answers) | USD 0.3125 OpenAI | USD 0.2475 lower SiliconFlow |
| 1 : 1 | USD 0.225 OpenAI | USD 0.195 lower SiliconFlow |
| 3 : 1 (the site’s blend) | USD 0.1375 lower OpenAI | USD 0.1425 SiliconFlow |
| 10 : 1 (retrieval, long prompts) | USD 0.08182 lower OpenAI | USD 0.1091 SiliconFlow |
| 30 : 1 (agents, big contexts) | USD 0.06129 lower OpenAI | USD 0.09677 SiliconFlow |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do GPT-5 Nano and Qwen3-30B-A3B-2507 cost per character in other languages?
Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the GPT (o200k) tokenizer for GPT-5 Nano and the Qwen3 32B tokenizer for Qwen3-30B-A3B-2507:
| Content | GPT-5 Nano | Qwen3-30B-A3B-2507 |
|---|---|---|
| English | USD 0.011 lower Microsoft Azure | USD 0.020 SiliconFlow |
| Arabic MSA | USD 0.016 lower Microsoft Azure | USD 0.035 SiliconFlow |
| Egyptian Arabic | USD 0.017 lower Microsoft Azure | USD 0.036 SiliconFlow |
| Hindi | USD 0.016 lower Microsoft Azure | USD 0.084 SiliconFlow |
| Chinese | USD 0.039 lower Microsoft Azure | USD 0.058 SiliconFlow |
| Spanish | USD 0.011 lower Microsoft Azure | USD 0.024 SiliconFlow |
| Code | USD 0.013 lower Microsoft Azure | USD 0.023 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
- GPT-5 Nano API pricing: all 3 providers and other tiers.
- Qwen3-30B-A3B-Instruct-2507 API pricing: all 2 providers and other tiers.
- Cost calculator with GPT-5 Nano and Qwen3-30B-A3B-2507 preloaded
- GPT token counter
- Qwen3 token counter
- Prompt caching calculator
More comparisons: Gemma 4 26B A4B vs GPT-5 Nano · GPT-5.4 Nano vs GPT-5 Nano · GPT-5 Mini vs GPT-5 Nano · GPT-5 Nano vs GPT-5 · Gemma 4 26B A4B vs Qwen3-30B-A3B-2507 · Qwen3-30B-A3B-2507 vs Qwen3 30B A3B · every GPT-5 Nano comparison · every Qwen3-30B-A3B-2507 comparison · all 662 price comparisons
Questions
Is GPT-5 Nano cheaper than Qwen3-30B-A3B-2507?
Yes. On a blend of 3 input tokens to 1 output token, GPT-5 Nano costs USD 0.1375 per 1M tokens and Qwen3-30B-A3B-2507 USD 0.1425, at each model's cheapest standard price. For input alone, GPT-5 Nano is cheaper (USD 0.05 against USD 0.09 per 1M); for output alone, Qwen3-30B-A3B-2507 is (USD 0.30 against USD 0.40). This compares price only, not quality.
What do GPT-5 Nano and Qwen3-30B-A3B-2507 cost per 1M tokens?
GPT-5 Nano: from USD 0.05 input and USD 0.40 output at OpenAI and 2 others; OpenAI's own API charges USD 0.05 / USD 0.40; median of 3 standard prices USD 0.05 / USD 0.40. Qwen3-30B-A3B-2507: from USD 0.09 input and USD 0.30 output at SiliconFlow; median of 2 standard prices USD 0.1075 / USD 0.375. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both GPT-5 Nano and Qwen3-30B-A3B-2507?
None of the providers we track sells both at a standard price. GPT-5 Nano is cheapest at OpenAI and 2 others and Qwen3-30B-A3B-2507 at SiliconFlow, so running both means two accounts.
Which is cheaper for a coding agent or a chatbot, GPT-5 Nano or Qwen3-30B-A3B-2507?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 57.15 a month on GPT-5 Nano (OpenAI) against USD 189.00 on Qwen3-30B-A3B-2507 (SiliconFlow), a difference of USD 131.85. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 9.27 a month on GPT-5 Nano (OpenAI) against USD 13.50 on Qwen3-30B-A3B-2507 (SiliconFlow), a difference of USD 4.23. 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 GPT-5 Nano and Qwen3-30B-A3B-2507 change?
The cheaper model changes with the mix: below 2.5 input tokens per output token, Qwen3-30B-A3B-2507 is cheaper; above 2.5, GPT-5 Nano is cheaper. At the site's 3:1 blend, GPT-5 Nano 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 GPT-5 Nano and Qwen3-30B-A3B-2507 have cached input pricing?
GPT-5 Nano from USD 0.005 per 1M cached input tokens (OpenAI; 3 of 3 standard prices list a discount). Qwen3-30B-A3B-2507: no standard price lists a cached-input discount. Cached tokens are billed at the input price where no cache price is published.