Google Gemini 2.5 Pro vs Xiaomi MiMo-V2-Pro — API Pricing, 150K Cache & Latency Benchmark
Side-by-side unit economics audit comparing Google Gemini 2.5 Pro (google-gemini-2-5-pro: Input $1.25/1M, Output $10/1M, Warm Cache Read $0.3125/1M) against Xiaomi MiMo-V2-Pro (xiaomi/mimo-v2-pro: Input $0.4/1M, Output $1.6/1M, Warm Cache Read $0.04/1M).
Activate 4-breakpoint prefix caching on Google Gemini 2.5 Pro to cut input token costs from $1.25/1M down to $0.3125/1M, or route async workloads via Batch API for 50% off.
Lock static system prompt prefixes on Xiaomi MiMo-V2-Pro to pay $0.04/1M warm cache rates instead of $0.4/1M uncached input, plus eligible startup cloud credits.
In Etuki FinOps telemetry comparing Google Gemini 2.5 Pro (API ID: google-gemini-2-5-pro) vs Xiaomi MiMo-V2-Pro (API ID: xiaomi/mimo-v2-pro), Xiaomi MiMo-V2-Pro offers lower standard uncached input pricing at $0.4/1M tokens, while Xiaomi MiMo-V2-Pro leads in 10-turn 150K cache-pinned coding economics at $0.1460 per 10 turns (saving $0.6634 over Google Gemini 2.5 Pro). For latency-sensitive interactive agents, Xiaomi MiMo-V2-Pro delivers 68ms faster P90 time-to-first-token (492ms vs 560ms).
| Specification / FinOps Dimension | Google Gemini 2.5 Pro (google-gemini-2-5-pro) | Xiaomi MiMo-V2-Pro (xiaomi/mimo-v2-pro) | Verified Advantage |
|---|---|---|---|
| Official Vendor API ID | google-gemini-2-5-pro | xiaomi/mimo-v2-pro | UMCE 563-Registry Locked |
| 563-Registry & Routing Eligibility | is_known: true | routable: true | is_known: true | routable: true | Both Merkle Verified |
| Standard Uncached Input (/1M tokens) | $1.25 / 1M | $0.4 / 1M | Xiaomi MiMo-V2-Pro |
| Warm Prefix Cache Read (/1M tokens) | $0.3125 / 1M | $0.04 / 1M | Xiaomi MiMo-V2-Pro |
| Standard Output (/1M tokens) | $10 / 1M | $1.6 / 1M | Xiaomi MiMo-V2-Pro |
| Context Window Capacity | 1049K Tokens | 256K Tokens | Google Gemini 2.5 Pro |
| 150K Prompt × 10 Turns (Unpinned Cold + 2K Out) | $2.0750 | $0.6320 | Xiaomi MiMo-V2-Pro |
| 150K × 10 Turns Case A (100% Static Pinned) | $0.8094 (saves $1.2656) | $0.1460 (saves $0.4860) | Xiaomi MiMo-V2-Pro (-$0.6634) |
| 150K × 10 Turns Case B (90% Static + 10% Tail) | $0.9359 (saves $1.1391) | $0.1946 (saves $0.4374) | Xiaomi MiMo-V2-Pro |
| P50 / P90 First-Token Latency | 381ms / 560ms (86 tok/s) | 335ms / 492ms (134 tok/s) | Xiaomi MiMo-V2-Pro (-68ms) |
1. Full-Turn Rate-Card Substitution for Google Gemini 2.5 Pro: At an uncached input rate of $1.25/1M, output rate of $10/1M, and warm cache read rate of $0.3125/1M, running 10 consecutive 150,000-token coding agent turns (with 2,000 output tokens/turn) without prefix pinning costs $2.0750. Locking 100% static prefixes (Case A: 1 cold turn + 9 warm cached turns + 20K total output) reduces 10-turn spend to $0.8094 (saving $1.2656); under a 90% static prefix + 10% dynamic tail workload (Case B), 10-turn spend is $0.9359.
2. Full-Turn Rate-Card Substitution for Xiaomi MiMo-V2-Pro & Winner Verdict: On Xiaomi MiMo-V2-Pro ($0.4/1M input, $1.6/1M output, $0.04/1M warm cache), 10 unpinned 150K turns cost $0.6320, dropping to $0.1460 in Case A and $0.1946 in Case B. Across every 10-turn 150K coding loop (including output tokens), Xiaomi MiMo-V2-Pro delivers an additional $0.6634 Case-A cost advantage.
Production Workload TCO: 25,000 MAU • 50M Tokens/Mo (50% Cache Hit) • 2,000 GB Egress
💡 Optimal FinOps Verdict: Deploying Xiaomi MiMo-V2-Pro saves $106.45/month ($1277/year) at 25,000 MAU & 50M monthly tokens (80% input / 20% output split with 50% prefix cache hit rate).