MiniMax-M3 API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of MiniMax-M3 for your use-case.
Lowest Time per Task
CoreWeave (NVFP4)
Fireworks
Nebius (FP8)
SiliconFlow (FP8)
MiniMaxTime per Intelligence Index task (minutes)
Total 10 providers
Lowest Cost per Task
GMI
DeepInfraSiliconFlow (FP8)
Modular
NovitaCost per Intelligence Index task
Total 10 providers
MiniMax-M3 is available through 10 API providers, each offering different performance characteristics and pricing. Below is a comparison of the key metrics across providers.
- For time per task, CoreWeave (NVFP4) (5.8 min), Fireworks (6.8 min), and Nebius (FP8) (8.3 min) finish an Intelligence Index task fastest.
- For cost per task, GMI ($0.91), DeepInfra ($1.11), and SiliconFlow (FP8) ($1.14) are the cheapest per Intelligence Index task.
Pricing
Cost per Intelligence Index Task: MiniMax-M3
Pricing: Cache Hit, Input, and Output: MiniMax-M3
Output Speed vs. Cost per Task: MiniMax-M3
Speed
Average of the single-request and multi-turn agentic workloads measured
Time per Intelligence Index Task: MiniMax-M3
Latency
Average of the single-request and multi-turn agentic workloads measured
Time to First Answer Token: MiniMax-M3
Reliability
Share of requests the endpoint failed, across all the single-request and multi-turn agentic workloads measured
Endpoint Error Rate: MiniMax-M3
Caching
Cache Hit Rate: MiniMax-M3
Cost per Task vs. Cache Hit Rate: MiniMax-M3
Key Comparison Metrics & API Features
| CoreWeave | $1.48 | 5.8 | 4.08 | 0.0% | 262k | Open | ||
| DeepInfra | $1.11 | 41.9 | 38.17 | 3.7% | 524k | Open | ||
| Fireworks | $1.26 | 6.8 | 6.24 | 0.0% | 524k | Open | ||
| GMI | $0.91 | 15.3 | 1.75 | 0.0% | 1M | Open | ||
| MiniMax | $1.29 | 14.2 | 71.75 | 0.0% | 512k | Open | ||
| Modular | $1.15 | 27.2 | 8.42 | 47.2% | 1M | Open | ||
| Nebius | $1.30 | 8.3 | 5.98 | 0.0% | 1M | Open | ||
| Novita | $1.16 | 14.6 | 67.74 | 0.0% | 1M | Open | ||
 Parasail | $1.17 | 15.0 | 6.12 | 12.7% | 1M | Open | ||
| SiliconFlow | $1.14 | 14.0 | 62.06 | 0.0% | 1M | Open | ||
Frequently Asked Questions
Common questions about MiniMax-M3 providers
MiniMax-M3 is available through 10 API providers: Parasail (MXFP8), CoreWeave (NVFP4), Novita, Modular, Nebius (FP8), SiliconFlow (FP8), MiniMax, DeepInfra, GMI, and Fireworks. Each provider offers different performance characteristics and pricing.
MiniMax-M3 is currently available through 10 API providers that we benchmark and track.
By time per task, the fastest providers for MiniMax-M3 are CoreWeave (NVFP4) (5.8 min), Fireworks (6.8 min), and Nebius (FP8) (8.3 min). Time per task divides the model's output tokens per Intelligence Index task by each provider's end-to-end output speed.
The providers with the lowest time to first answer token for MiniMax-M3 are GMI (1.75s), CoreWeave (NVFP4) (4.08s), and Nebius (FP8) (5.98s). Lower latency means faster initial response time.
The most affordable providers for MiniMax-M3 by blended price are CoreWeave (NVFP4) ($0.18 per 1M tokens), GMI ($0.18 per 1M tokens), and DeepInfra ($0.21 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-M3 are CoreWeave (NVFP4) ($0.23 per 1M input tokens), GMI ($0.24 per 1M input tokens), and DeepInfra ($0.28 per 1M input tokens).
The providers with the lowest output token pricing for MiniMax-M3 are CoreWeave (NVFP4) ($0.96 per 1M output tokens), GMI ($0.96 per 1M output tokens), and DeepInfra ($1.10 per 1M output tokens).
Prices for MiniMax-M3 vary up to 1.2x across providers. The most affordable is CoreWeave (NVFP4) at $0.18 per 1M tokens, while Parasail (MXFP8) charges $0.22 per 1M tokens.
Time per task for MiniMax-M3 varies significantly across providers. CoreWeave (NVFP4) is the quickest at 5.8 minutes per Intelligence Index task, 7.2x quicker than DeepInfra at 41.9 minutes.
All 10 providers of MiniMax-M3 support JSON mode for structured output.
All 10 providers of MiniMax-M3 support function calling (tool use).
The best provider for MiniMax-M3 depends on your priorities: CoreWeave (NVFP4) has the lowest time per task, GMI has the lowest latency, and CoreWeave (NVFP4) provides the most competitive pricing.
When choosing a provider for MiniMax-M3, consider: time per task (how quickly the provider completes an Intelligence Index task), 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-M3's intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview