As of August 20, 2026, Gemma 4 31B is 18% cheaper per million tokens than Hy3. Gemma 4 31B is $0.14 / $0.40 per million input/output tokens versus Hy3 at $0.13 / $0.53. Sources: https://ai.google.dev/gemini-api/docs/pricing and https://www.tencentcloud.com/document/product/1300/78937
Sources: Gemini API pricing (https://ai.google.dev/gemini-api/docs/pricing); Tencent pricing (https://www.tencentcloud.com/document/product/1300/78937)
Gemma 4 31B
Input
$0.14
Output
$0.40
Hy3
Tencent
Input
$0.13
Output
$0.53
Token pricing, specs, and benchmarks side by side
Gemma 4 31BHy3Verdict
Gemma 4 31B offers significantly lower pricing, while Hy3 leads on benchmark performance. Your choice depends on whether cost efficiency or raw capability matters more for your use case.
As of August 20, 2026, Gemma 4 31B is 18% cheaper on blended LLM API pricing. Gemma 4 31B costs $0.14 per million input tokens and $0.40 per million output tokens ($0.54 blended 1M-in + 1M-out). Hy3 costs $0.13 / $0.53 per million tokens ($0.66 blended). Sources: https://ai.google.dev/gemini-api/docs/pricing and https://www.tencentcloud.com/document/product/1300/78937. The cheapest LLM for your app still depends on how many output tokens you generate.
Gemma 4 31B Gemini API pricing is $0.14 input and $0.40 output per million tokens via Google. Hy3 Tencent API pricing is $0.13 input and $0.53 output per million tokens via Tencent. Use those four numbers, not ChatGPT Plus or Claude Pro subscription prices, when you are comparing APIs.
Hy3 has a published SWE-bench Verified score of 78%. Gemma 4 31B does not have that eval on this page yet.
Gemma 4 31B supports a 262K token context window, while Hy3 supports 262K tokens. Both models have the same context window size. Filling a larger window bills more input tokens, so the cheaper-per-million model can still cost more on long documents.
Hy3 leads on 2 of 2 shared benchmarks versus Gemma 4 31B's 0 wins. Check the comparison table for GPQA Diamond, SWE-bench, MMLU, HLE, and the other evals we track. Token price and benchmark score together are the usual LLM comparison, not either number alone.
Both Gemma 4 31B and Hy3 are production API models. For cost-sensitive production traffic, Gemma 4 31B has better economics on the blended 1M-in + 1M-out scale. For maximum capability, weight the benchmark table for your domain. Many production stacks route cheap models for drafts and a frontier model for the hard turn, which is usually cheaper than sending everything to the expensive API.
Yes. Call Gemma 4 31B through Google and Hy3 through Tencent with separate API keys, or through a gateway that already wraps both. Keep prompts in tokens, not characters, when you estimate the invoice. AnotherWrapper templates can swap providers without rewriting auth and billing.
List rates are the latest published Gemini API pricing and Tencent API pricing, quoted per million tokens. Official rate sheets: https://ai.google.dev/gemini-api/docs/pricing and https://www.tencentcloud.com/document/product/1300/78937. Prompt caching, batch APIs, and committed-use discounts can change the invoice. Benchmark scores are from official publications and independent evals, updated as new numbers land.
Gemma 4 31B supports up to 131K output tokens per request, while Hy3 supports up to an unspecified number of output tokens. Output tokens are usually the expensive half of API pricing, so a higher max output cap is a capability, not a discount.
Gemma 4 31B has a measured throughput of approximately 1.2030715399542904 tokens/second. Throughput data for Hy3 is not yet available. Throughput (tokens per second) and time-to-first-token are why two models with the same per-million-token price can still feel different in chat and agents.
Use the model pickers at the top of this LLM comparison. Search any model in the index and the URL updates. Open Gemma 4 31B or Hy3 from the sidebar for the single-model API pricing page, or go back to the LLM pricing table to sort by cheapest blended cost.
This comparison covers API pricing, context window, throughput, and benchmark scores. Gemma 4 31B costs $0.14/M input and $0.40/M output. Hy3 costs $0.13/M input and $0.53/M output.
Rank both on the LLM leaderboard, SWE-bench, and Humanity's Last Exam.
The index
280 models across 33 providers. Search or jump to a lab — every model page stays linked here.
Next step
Auth, billing, and the API layer are already decided. Fork 8 finished AI apps, or have us build the first version with you.
