Hunyuan Hy3 Preview vs Xiaomi MiMo-V2.5-Pro
Hunyuan Hy3 Preview (2026) and Xiaomi MiMo-V2.5-Pro (2026) compare a standalone API model against a coding-specialized model. Hunyuan Hy3 Preview ships a 262k-token context window, while Xiaomi MiMo-V2.5-Pro ships a 1.05m-token context window. On SWE-bench Verified, Xiaomi MiMo-V2.5-Pro leads by 4.5 pts. On pricing, Hunyuan Hy3 Preview costs $0.07/1M input tokens versus $0.43/1M for the alternative. This page treats the result as workflow and deployment fit, not a universal model winner.
Treat this as a product-type comparison: Hunyuan Hy3 Preview is standalone API model, while Xiaomi MiMo-V2.5-Pro is coding-specialized model. Choose based on workflow fit before reading any benchmark or price row as decisive.
Decision scorecard
Local evidence first| Signal | Hunyuan Hy3 Preview | Xiaomi MiMo-V2.5-Pro |
|---|---|---|
| Product type | Standalone API model | Coding-specialized model |
| Best for | reasoning-heavy apps, tool-calling agents, and provider-routed production | custom coding agents, code generation, and tool loops |
| Decision fit | Coding, RAG, and Agents | Coding, RAG, and Agents |
| Context window | 262k | 1.05m |
| Cheapest output | $0.26/1M tokens | $0.87/1M tokens |
| Provider routes | 2 tracked | 3 tracked |
| Shared benchmarks | 3 rows | SWE-bench Verified leader |
Decision tradeoffs
- Hunyuan Hy3 Preview holds a shared-benchmark lead on Google-Proof Q&A, ahead by 20.5 points.
- Hunyuan Hy3 Preview has the lower cheapest tracked output price at $0.26/1M tokens.
- Hunyuan Hy3 Preview uniquely exposes Reasoning in local model data.
- Local decision data tags Hunyuan Hy3 Preview for Coding, RAG, and Agents.
- Xiaomi MiMo-V2.5-Pro holds a shared-benchmark lead on SWE-bench Verified, ahead by 4.5 points.
- Xiaomi MiMo-V2.5-Pro has the larger context window for long prompts, retrieval packs, or transcript analysis.
- Xiaomi MiMo-V2.5-Pro has broader tracked provider coverage for fallback and procurement flexibility.
- Xiaomi MiMo-V2.5-Pro uniquely exposes Structured outputs in local model data.
- Local decision data tags Xiaomi MiMo-V2.5-Pro for Coding, RAG, and Agents.
Monthly cost at traffic
Estimate token spend from the cheapest tracked input and output route or tier on this page.
Hunyuan Hy3 Preview
$118
Cheapest tracked route/tier: OpenRouter
Xiaomi MiMo-V2.5-Pro
$566
Cheapest tracked route/tier: OpenRouter
Estimated monthly gap: $448. Batch, cache, alternate speed tiers, and negotiated pricing are excluded from this local estimate.
Switch friction
- Provider overlap exists on OpenRouter; start route-level A/B tests there.
- Xiaomi MiMo-V2.5-Pro is $0.61/1M tokens higher on cheapest tracked output pricing, so quality gains need to justify the spend.
- Check replacement coverage for Reasoning before moving production traffic.
- Xiaomi MiMo-V2.5-Pro adds Structured outputs in local capability data.
- Provider overlap exists on OpenRouter; start route-level A/B tests there.
- Hunyuan Hy3 Preview is $0.61/1M tokens lower on cheapest tracked output pricing before cache, batch, or negotiated discounts.
- Check replacement coverage for Structured outputs before moving production traffic.
- Hunyuan Hy3 Preview adds Reasoning in local capability data.
Specs
| Specification | ||
|---|---|---|
| Released | 2026-04-22 | 2026-04-22 |
| Context window | 262k | 1.05m |
| Parameters | 295B | 1T |
| Architecture | dense moe hybrid | mixture of experts |
| License | Tencent Hunyuan Community License | Proprietary |
| Openness | Open weights | Proprietary |
| Commercial use | Commercial use with conditions | Commercial use with conditions |
| Knowledge cutoff | - | - |
Pricing and availability
| Pricing attribute | Hunyuan Hy3 Preview | Xiaomi MiMo-V2.5-Pro |
|---|---|---|
| Input price | $0.07/1M tokens | $0.43/1M tokens |
| Output price | $0.26/1M tokens | $0.87/1M tokens |
| Providers |
Capabilities
| Capability | Hunyuan Hy3 Preview | Xiaomi MiMo-V2.5-Pro |
|---|---|---|
| Vision | No | No |
| Multimodal | No | No |
| Reasoning | Yes | No |
| Function calling | Yes | Yes |
| Tool use | Yes | Yes |
| Structured outputs | No | Yes |
| Code execution | No | No |
| IDE integration | No | No |
| Computer use | No | No |
| Parallel agents | No | No |
Benchmarks
| Benchmark | Hunyuan Hy3 Preview | Xiaomi MiMo-V2.5-Pro |
|---|---|---|
| SWE-bench Verified | 74.4 | 78.9 |
| Google-Proof Q&A | 87.2 | 66.7 |
| Terminal-Bench 2.0 | 54.4 | 68.4 |
Deep dive
On shared benchmark coverage, SWE-bench Verified has Hunyuan Hy3 Preview at 74.4 and Xiaomi MiMo-V2.5-Pro at 78.9, with Xiaomi MiMo-V2.5-Pro ahead by 4.5 points; Google-Proof Q&A has Hunyuan Hy3 Preview at 87.2 and Xiaomi MiMo-V2.5-Pro at 66.7, with Hunyuan Hy3 Preview ahead by 20.5 points; Terminal-Bench 2.0 has Hunyuan Hy3 Preview at 54.4 and Xiaomi MiMo-V2.5-Pro at 68.4, with Xiaomi MiMo-V2.5-Pro ahead by 14.0 points. The largest visible gap is 20.5 points on Google-Proof Q&A, which matters most when that benchmark mirrors your workload. Treat isolated benchmark wins as directional, because provider routing, prompt style, and tool access can move real application results.
The capability footprint differs most on reasoning mode: Hunyuan Hy3 Preview and structured outputs: Xiaomi MiMo-V2.5-Pro. Both models share function calling and tool use, so the practical split is not just feature count. Use those differences to decide whether the page is about raw model quality, agentic coding support, multimodal ingestion, or predictable structured API behavior.
For cost, Hunyuan Hy3 Preview lists $0.07/1M input and $0.26/1M output tokens on the cheapest tracked provider, while Xiaomi MiMo-V2.5-Pro lists $0.43/1M input and $0.87/1M output tokens on the cheapest tracked provider. A 70/30 input-output blend puts Hunyuan Hy3 Preview lower by about $0.44 per million blended tokens. Availability is 2 providers versus 3, so concentration risk also matters.
Choose Hunyuan Hy3 Preview when reasoning depth and lower input-token cost are central to the workload. Choose Xiaomi MiMo-V2.5-Pro when coding workflow support, larger context windows, and broader provider choice are more important. For production, rerun your own prompts through the exact provider, region, and tool stack you plan to ship.
FAQ
Which has a larger context window, Hunyuan Hy3 Preview or Xiaomi MiMo-V2.5-Pro?
Xiaomi MiMo-V2.5-Pro supports 1.05m tokens, while Hunyuan Hy3 Preview supports 262k tokens. That gap matters most for long documents, large codebases, retrieval-heavy agents, and conversations where earlier context must remain visible.
Which is cheaper, Hunyuan Hy3 Preview or Xiaomi MiMo-V2.5-Pro?
Hunyuan Hy3 Preview is cheaper on tracked token pricing. Hunyuan Hy3 Preview costs $0.07/1M input and $0.26/1M output tokens. Xiaomi MiMo-V2.5-Pro costs $0.43/1M input and $0.87/1M output tokens. Provider discounts or batch pricing can still change the final bill.
Is Hunyuan Hy3 Preview or Xiaomi MiMo-V2.5-Pro open source?
Hunyuan Hy3 Preview is listed under Tencent Hunyuan Community License. Xiaomi MiMo-V2.5-Pro is listed under Proprietary. License labels affect whether you can self-host, redistribute weights, or rely only on hosted APIs, so confirm the upstream license before deployment.
Which is better for reasoning mode, Hunyuan Hy3 Preview or Xiaomi MiMo-V2.5-Pro?
Hunyuan Hy3 Preview has the clearer documented reasoning mode signal in this comparison. If reasoning mode is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Which is better for function calling, Hunyuan Hy3 Preview or Xiaomi MiMo-V2.5-Pro?
Both Hunyuan Hy3 Preview and Xiaomi MiMo-V2.5-Pro expose function calling. The better choice depends on benchmark fit, context budget, pricing, and whether your provider route exposes the same capability surface.
Where can I run Hunyuan Hy3 Preview and Xiaomi MiMo-V2.5-Pro?
Hunyuan Hy3 Preview is available on OpenRouter and Tencent Cloud TI Platform. Xiaomi MiMo-V2.5-Pro is available on OpenRouter, Xiaomi, and Novita AI. Provider coverage can affect latency, region availability, compliance posture, and fallback options.
Continue comparing
Last reviewed: 2026-06-07. Data sourced from public model cards and provider documentation.