Tencentopen weightsHy4 preview pricing
$0.83 per 1M input tokens and $2.50 per 1M output tokens. In real work that is 0.35¢ per chat answer and 1.7¢ to summarize a 20-page PDF.
What Hy4 preview costs per task
Token counts estimated from a measured model of the same lab (Mistral Large 4), rescaled by tokenizer. Reasoning tokens are billed as output.
| Task | Cost | Tokens used | Price rank | Basis |
|---|---|---|---|---|
| Answer a chat question | 0.35¢$3.48 / 1,000 | 40 in · 523 out · 854 reasoning | 205 of 336 | estimated |
| Reply to a customer email | 0.91¢$9.08 / 1,000 | 254 in · 599 out · 2,946 reasoning | 303 of 336 | estimated |
| Write a 1,500-word blog post | 1.9¢$19.48 / 1,000 | 55 in · 2,064 out · 5,705 reasoning | 236 of 336 | estimated |
| Write 10 product descriptions | 0.79¢$7.94 / 1,000 | 285 in · 849 out · 2,230 reasoning | 188 of 336 | estimated |
| Translate 1,000 words | 3.1¢$31.47 / 1,000 | 1,349 in · 1,317 out · 10,816 reasoning | 277 of 336 | estimated |
| Proofread one page (500 words) | 1.1¢$10.97 / 1,000 | 731 in · 860 out · 3,281 reasoning | 271 of 336 | estimated |
| Summarize a 20-page PDF | 1.7¢$16.58 / 1,000 | 13,373 in · 151 out · 2,019 reasoning | 235 of 336 | estimated |
| Extract data from an invoice | 0.70¢$6.95 / 1,000 | 374 in · 871 out · 1,784 reasoning | 259 of 336 | estimated |
| Analyze a 200-row spreadsheet | 3.1¢$30.74 / 1,000 | 4,774 in · 4,526 out · 6,174 reasoning | 192 of 336 | estimated |
| Answer from 5 retrieved documents (RAG) | 1.1¢$11.13 / 1,000 | 3,772 in · 210 out · 2,981 reasoning | 244 of 336 | estimated |
| Classify 100 support tickets | 2.0¢$19.56 / 1,000 | 1,274 in · 832 out · 6,565 reasoning | 258 of 336 | estimated |
| Review 500 lines of code | 2.7¢$26.68 / 1,000 | 4,658 in · 1,594 out · 7,522 reasoning | 220 of 336 | estimated |
| Write a SQL query | 0.80¢$8.00 / 1,000 | 121 in · 511 out · 2,647 reasoning | 265 of 336 | estimated |
| Run a 50-step coding agent | 53¢$533 / 1,000 | 2,304,625 in · 15,450 out · 132,350 reasoning | 181 of 336 | modeled |
Price at every provider
List prices per 1M tokens (input / output) at each host serving Hy4 preview through OpenRouter today. Hosts can differ in quantization, context and speed.
| Provider | Input / output | Cached input | Context | Uptime |
|---|---|---|---|---|
 DeepInfrafp4 | $0.83 / $2.50 | $0.042 | 1M | 100% |
 Novitafp8 | $0.83 / $2.50 | $0.042 | 1M | 100% |
 SiliconFlowfp8 | $0.83 / $2.50 | $0.042 | 1M | 100% |
| T Tencentfp8 | $0.83 / $2.50 | $0.042 | 1M | 100% |
Price history
First seen in our price history on Sep 2, 2026 at $0.83 input / $2.50 output. Since then the output price moved from $2.50 to $2.50 per 1M tokens (2 changes).
FAQ
How much does Hy4 preview cost?
Hy4 preview costs $0.83 per 1 million input tokens and $2.50 per 1 million output tokens, with cached input at $0.042 (list price on Oct 18, 2026). In practice one chat answer costs about 0.35¢ and summarizing a 20-page PDF about 1.7¢.
Where is Hy4 preview cheapest?
Among 4 providers we track, the lowest list price today is at DeepInfra: $0.83 input and $2.50 output per 1M tokens.
Does Hy4 preview charge for reasoning tokens?
Yes. Reasoning (thinking) tokens are billed as output tokens. That is why the cost per task can be much higher than the visible answer suggests - our task costs include them.
Is the Hy4 preview cost per task measured or estimated?
Estimated: Hy4 preview was not run directly. Tokens come from Mistral Large 4 and are rescaled by the tokenizer of Tencent. Real usage can differ.