Hy4 vs Hy3 API pricing
Updated
Hy4: USD 0.417 input, USD 1.2505 output per 1M tokens (GMI Cloud and 1 other, promotional). Hy3: USD 0.132 input, USD 0.528 output per 1M tokens (SiliconFlow). At 3 input tokens per output token, Hy4 costs 2.71× as much as Hy3 (USD 0.6254 vs USD 0.231 per 1M).
How do Hy4 and Hy3 differ on price?
- Cheapest input price: Hy3 USD 0.132 against USD 0.417 per 1M tokens, so Hy4 costs 3.2× as much.
- Cheapest output price: Hy3 USD 0.528 against USD 1.2505 per 1M tokens, so Hy4 costs 2.4× as much.
- Among the 198 indexed models, Hy4 ranks 105th cheapest and Hy3 59th on the 3:1 blend.
- No input:output mix changes the answer: Hy3 is cheaper at every mix before cache discounts.
- Both list cached-input discounts: Hy4 at 3 standard prices of 3, from USD 0.021 (GMI Cloud, 95% below that provider's input price); Hy3 at 3 standard prices of 4, from USD 0.033 (SiliconFlow, 75% below that provider's input price).
- Hy4 states up to 1,048,576 tokens of context, 4× the 262,144 of Hy3.
- Hy3 has 4 providers with a standard price, 1 more than Hy4; 2 providers sell both (GMI Cloud and Novita AI).
- On one account, GMI Cloud charges USD 0.6254 for Hy4 (its cheapest price) and USD 0.25 for Hy3 (1.08× its cheapest price) at 3:1, tied with 1 other provider for the lowest combined price among the providers that sell both.
- Shopping around matters more for Hy3: its dearest standard price is 7.6× its cheapest, against 1.25× for Hy4.
- For the one-developer coding agent preset, the gap is USD 1,483 a year in Hy3's favour (Hy4 USD 257.26 a month, Hy3 USD 133.65).
Hy4 and Hy3 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 | Hy4 | Hy3 |
|---|---|---|
| Model page | Hy4 Tencent · 3 providers | Hy3 Tencent · 4 providers |
| Cheapest standard price, blended 3:1 | USD 0.6254 promo USD 0.417 in / USD 1.2505 out at GMI Cloud and 1 other GMI Cloud page checked 2026-11-15 | USD 0.231 lower USD 0.132 in / USD 0.528 out at SiliconFlow SiliconFlow page checked 2026-11-14 |
| Cheapest input | USD 0.417 promo GMI Cloud source | USD 0.132 lower SiliconFlow source |
| Cheapest cached input | USD 0.021 lower promo GMI Cloud source | USD 0.033 SiliconFlow source |
| Cheapest output | USD 1.2505 promo GMI Cloud source | USD 0.528 lower SiliconFlow source |
| Cheapest non-promotional price | USD 0.525 / USD 1.5625 Chutes, USD 0.7844 blended source | The cheapest price is not promotional. |
| Maker's own API | Tencent publishes no paid standard price on its own API that we found. | Tencent publishes no paid standard price on its own API that we found. |
| Median standard price | USD 0.417 / USD 1.2505 input / output, 3 prices | USD 0.14 / USD 0.58 input / output, 4 prices |
| Context window | 1,048,576 tokens, largest stated; from 1,000,000 at some providers | 262,144 tokens, largest stated; from 32,768 at some providers |
| Providers with a standard price | 3 | 4 |
- Lower marks the lower price on a line; tie, equal prices. Other tiers and free rows are on the model pages.
- Promo: the provider calls the price promotional or discounted, so it is current now and can end without notice.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What do Hy4 and Hy3 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 | Hy4 | Hy3 | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 38.66 GMI Cloud | USD 16.04 SiliconFlow | USD 22.62 Hy3 lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 90.33 GMI Cloud | USD 32.47 SiliconFlow | USD 57.85 Hy3 lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 257.26 GMI Cloud | USD 133.65 SiliconFlow | USD 123.61 Hy3 lower |
Hy3 is cheaper on all three workloads. The widest gap is on RAG over documents: the dearer model costs 2.78× as much as the cheaper one.
Does the cheaper model change with your input:output ratio?
Hy3 is cheaper at every mix of input and output tokens, before cache discounts: its cheapest input price and its cheapest output price are both at or below Hy4's, so the input:output ratio never changes the answer.
| Input : output tokens | Hy4 | Hy3 |
|---|---|---|
| 1 : 3 (long answers) | USD 1.042 GMI Cloud | USD 0.429 lower SiliconFlow |
| 1 : 1 | USD 0.8337 GMI Cloud | USD 0.33 lower SiliconFlow |
| 3 : 1 (the site’s blend) | USD 0.6254 GMI Cloud | USD 0.231 lower SiliconFlow |
| 10 : 1 (retrieval, long prompts) | USD 0.4928 GMI Cloud | USD 0.168 lower SiliconFlow |
| 30 : 1 (agents, big contexts) | USD 0.4439 GMI Cloud | USD 0.1448 lower SiliconFlow |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do Hy4 and Hy3 cost per character in other languages?
Not measured: the real cost by language dataset covers only tokenizers we could measure from published files, and Hy4 and Hy3 are not among them. Count tokens on your own text with the counters below.
Check it on your own traffic
- Hy4 API pricing: all 3 providers and other tiers.
- Hy3 API pricing: all 4 providers and other tiers.
- Cost calculator with Hy4 and Hy3 preloaded
- Token counter
- Prompt caching calculator
Questions
Is Hy4 cheaper than Hy3?
No. On a blend of 3 input tokens to 1 output token, Hy3 costs USD 0.231 per 1M tokens and Hy4 USD 0.6254, at each model's cheapest standard price. For input alone, Hy3 is cheaper (USD 0.132 against USD 0.417 per 1M); for output alone, Hy3 is (USD 0.528 against USD 1.2505). This compares price only, not quality.
What do Hy4 and Hy3 cost per 1M tokens?
Hy4: from USD 0.417 input and USD 1.2505 output at GMI Cloud and 1 other (promotional); median of 3 standard prices USD 0.417 / USD 1.2505. Hy3: from USD 0.132 input and USD 0.528 output at SiliconFlow; median of 4 standard prices USD 0.14 / USD 0.58. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both Hy4 and Hy3?
2 providers have a standard price for both: GMI Cloud (USD 0.6254 and USD 0.25 at 3:1), Novita AI (USD 0.6254 and USD 0.25 at 3:1). The first 2 tie for the lowest combined price. Buying each model from its own cheapest provider costs USD 0.6254 and USD 0.231.
Which is cheaper for a coding agent or a chatbot, Hy4 or Hy3?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 257.26 a month on Hy4 (GMI Cloud) against USD 133.65 on Hy3 (SiliconFlow), a difference of USD 123.61. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 38.66 a month on Hy4 (GMI Cloud) against USD 16.04 on Hy3 (SiliconFlow), a difference of USD 22.62. 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 Hy4 and Hy3 change?
Hy3 is cheaper at every mix of input and output tokens, before cache discounts: its cheapest input price and its cheapest output price are both at or below Hy4's, so the input:output ratio never changes the answer.
Do Hy4 and Hy3 have cached input pricing?
Hy4 from USD 0.021 per 1M cached input tokens (GMI Cloud; 3 of 3 standard prices list a discount). Hy3 from USD 0.033 per 1M cached input tokens (SiliconFlow; 3 of 4 standard prices list a discount). Cached tokens are billed at the input price where no cache price is published.
Which has the larger context window, Hy4 or Hy3?
Hy4: up to 1,048,576 tokens, against 262,144 for Hy3, 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.