Gemma 4 31B (Reasoning) API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Gemma 4 31B (Reasoning) for your use-case.
Fastest
Cerebras
SambaNova
Modular (NVFP4)
FriendliAI
CoreWeaveOutput speed
Total 14 providers
Lowest Latency
CerebrasModular (NVFP4)SambaNovaFriendliAICoreWeaveTime to first answer token
Total 14 providers
Lowest Price
Google (AI Studio)
DeepInfra
ParasailCoreWeave
SiliconFlow (FP8)Blended price (per 1M tokens)
Total 14 providers
Gemma 4 31B is available through 14 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 Cerebras (1,388.6 t/s), SambaNova (198.4 t/s), and Modular (NVFP4) (175.2 t/s). Speed varies significantly across providers, with a 7515% difference between the fastest and slowest.
- For latency, Cerebras (1.70s), Modular (NVFP4) (10.38s), and SambaNova (11.28s) offer the lowest time to first answer token.
- For pricing, Google (AI Studio) (0.00), DeepInfra (0.08), and Parasail (0.11) offer the lowest blended prices per 1M tokens.
- Cerebras offers the best performance with both the highest speed and lowest latency. For cost optimization, Google (AI Studio) provides the most competitive pricing.
Highlights
Update: Default performance benchmarking workload has updated to 10k input tokens to better reflect production use cases. You can still select different workloads above.
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: Gemma 4 31B (Reasoning)
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
| CoreWeave | 262k | Open | $0.07 | 56 | 2.02 | 41.98 | 31.03 | ||
| Modular | 262k | Open | -- | 175 | 0.47 | 13.23 | 9.91 | ||
 GMI | 262k | Open | $0.10 | -- | -- | -- | -- | ||
 Novita | 262k | Open | $0.10 | -- | -- | -- | -- | ||
| Cerebras | 131k | Open | $0.71 | 1,389 | 0.45 | 2.06 | 1.25 | ||
| Google | 262k | Open | $0.00 | 35 | 1.11 | 64.50 | 49.22 | ||
| Parasail | 262k | Open | $0.05 | 24 | 1.43 | 94.93 | 72.59 | ||
| DeepInfra | 262k | Open | $0.02 | 25 | 1.49 | 90.68 | 69.24 | ||
 Crusoe | 262k | Open | $0.10 | 26 | 1.44 | 86.53 | 66.06 | ||
| SiliconFlow | 262k | Open | $0.10 | 44 | 3.88 | 55.27 | 39.90 | ||
| SambaNova | 131k | Open | $0.28 | 198 | 2.53 | 13.80 | 8.75 | ||
| FriendliAI | 256k | Open | $0.10 | 104 | 1.70 | 23.21 | 16.70 | ||
 Lightning AI | 131k | Open | $0.10 | -- | -- | -- | -- | ||
 Together AI | 262k | Open | $0.29 | 18 | 1.38 | 124.00 | 95.20 | ||
Frequently Asked Questions
Common questions about Gemma 4 31B (Reasoning) providers
Gemma 4 31B (Reasoning) is available through 14 API providers: CoreWeave, Modular (NVFP4), GMI (FP8), Novita, Cerebras, Google (AI Studio), Parasail, DeepInfra, Crusoe, SiliconFlow (FP8), SambaNova, FriendliAI, Lightning AI, and Together AI (FP8). Each provider offers different performance characteristics and pricing.
Gemma 4 31B (Reasoning) is currently available through 14 API providers that we benchmark and track.
The fastest providers for Gemma 4 31B (Reasoning) by output speed are Cerebras (1,388.6 t/s), SambaNova (198.4 t/s), and Modular (NVFP4) (175.2 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 Gemma 4 31B (Reasoning) are Cerebras (1.70s), Modular (NVFP4) (10.38s), and SambaNova (11.28s). Lower latency means faster initial response time.
The most affordable providers for Gemma 4 31B (Reasoning) by blended price are Google (AI Studio) ($0.00 per 1M tokens), DeepInfra ($0.08 per 1M tokens), and Parasail ($0.11 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 Gemma 4 31B (Reasoning) are Google (AI Studio) ($0.00 per 1M input tokens), CoreWeave ($0.10 per 1M input tokens), and DeepInfra ($0.13 per 1M input tokens).
The providers with the lowest output token pricing for Gemma 4 31B (Reasoning) are Google (AI Studio) ($0.00 per 1M output tokens), CoreWeave ($0.34 per 1M output tokens), and DeepInfra ($0.38 per 1M output tokens).
Output speed for Gemma 4 31B (Reasoning) varies significantly across providers. Cerebras is the fastest at 1,388.6 t/s, which is 76.2x faster than Together AI (FP8) at 18.2 t/s.
All 14 providers of Gemma 4 31B (Reasoning) support JSON mode for structured output.
All 14 providers of Gemma 4 31B (Reasoning) support function calling (tool use).
For Gemma 4 31B (Reasoning), Cerebras offers the best performance with highest speed and lowest latency. For cost optimization, Google (AI Studio) provides the most competitive pricing.
When choosing a provider for Gemma 4 31B (Reasoning), 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 Gemma 4 31B (Reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview