This model is deprecated. We only continue performance benchmarking for the default 10k input token workload. Results for other workloads are historical and no longer updated.
MiniMax has launched a newer model, MiniMax-M2.7. We suggest considering it instead.
For more information, see comparison of MiniMax-M2.7 to other models and API provider benchmarks for MiniMax-M2.7.
MiniMax-M2.5 API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of MiniMax-M2.5 for your use-case.
Fastest
FriendliAI
Novita
GMI FP8
MiniMax
DigitalOceanOutput speed
Total 7 providers
Lowest Latency
FriendliAINovitaGMI FP8MiniMaxDigitalOceanTime to first answer token
Total 7 providers
Lowest Price
DeepInfra (FP8)NovitaMiniMaxGMI FP8
Together AI (FP4)Blended price (per 1M tokens)
Total 7 providers
MiniMax-M2.5 is available through 7 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.
- For output speed, the top providers are FriendliAI (149.7 t/s), Novita (91.7 t/s), and GMI FP8 (89.7 t/s). Speed varies significantly across providers, with a 609% difference between the fastest and slowest.
- For latency, FriendliAI (13.95s), Novita (23.62s), and GMI FP8 (24.38s) offer the lowest time to first answer token.
- For pricing, DeepInfra (FP8) (0.17), Novita (0.20), and MiniMax (0.20) offer the lowest blended prices per 1M tokens.
- FriendliAI offers the best performance with both the highest speed and lowest latency. For cost optimization, DeepInfra (FP8) provides the most competitive pricing.
Highlights
Pricing
Pricing: Cache Hit, Input, and Output
Pricing: Blended Price
Pricing: Cache Discount
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed: MiniMax-M2.5
Latency vs. Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First Answer Token
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time
Key Comparison Metrics & API Features
Frequently Asked Questions
Common questions about MiniMax-M2.5 providers
MiniMax-M2.5 is available through 7 API providers: Novita, MiniMax, GMI FP8, Together AI (FP4), DeepInfra (FP8), DigitalOcean, and FriendliAI. Each provider offers different performance characteristics and pricing.
MiniMax-M2.5 is currently available through 7 API providers that we benchmark and track.
The fastest providers for MiniMax-M2.5 by output speed are FriendliAI (149.7 t/s), Novita (91.7 t/s), and GMI FP8 (89.7 t/s). Output speed measures how quickly tokens are generated after the model starts responding.
The providers with the lowest time to first answer token for MiniMax-M2.5 are FriendliAI (13.95s), Novita (23.62s), and GMI FP8 (24.38s). Lower latency means faster initial response time.
The most affordable providers for MiniMax-M2.5 by blended price are DeepInfra (FP8) ($0.17 per 1M tokens), Novita ($0.20 per 1M tokens), and MiniMax ($0.20 per 1M tokens). Blended price uses a 7:2:1 cache hit/input/output token ratio.
The providers with the lowest input token pricing for MiniMax-M2.5 are DeepInfra (FP8) ($0.15 per 1M input tokens), Novita ($0.30 per 1M input tokens), and MiniMax ($0.30 per 1M input tokens).
The providers with the lowest output token pricing for MiniMax-M2.5 are DeepInfra (FP8) ($1.15 per 1M output tokens), Novita ($1.20 per 1M output tokens), and MiniMax ($1.20 per 1M output tokens).
Prices for MiniMax-M2.5 vary up to 1.3x across providers. The most affordable is DeepInfra (FP8) at $0.17 per 1M tokens, while Together AI (FP4) charges $0.22 per 1M tokens.
Output speed for MiniMax-M2.5 varies significantly across providers. FriendliAI is the fastest at 149.7 t/s, which is 7.1x faster than DeepInfra (FP8) at 21.1 t/s.
All 7 providers of MiniMax-M2.5 support JSON mode for structured output.
All 7 providers of MiniMax-M2.5 support function calling (tool use).
For MiniMax-M2.5, FriendliAI offers the best performance with highest speed and lowest latency. For cost optimization, DeepInfra (FP8) provides the most competitive pricing.
When choosing a provider for MiniMax-M2.5, consider: output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and API features like JSON mode or function calling.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
For information about MiniMax-M2.5's intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview