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 Results for other workloads are historical and no longer updated.
IBM has launched a newer model, Granite 4.0 Micro. We suggest considering it instead.
For more information, see comparison of Granite 4.0 Micro to other models and API provider benchmarks for Granite 4.0 Micro.
Granite 3.3 8B (Non-reasoning) API Provider Benchmarking & Analysis
This analysis is intended to support you in choosing the best API provider of Granite 3.3 8B (Non-reasoning) for your use-case.
Fastest
ReplicateOutput speed
Total 1 provider
Lowest Latency
ReplicateTime to first token
Total 1 provider
Lowest Price
ReplicateBlended price (per 1M tokens)
Total 1 provider
Granite 3.3 8B is available through Replicate.
Highlights
Pricing
Pricing: Cache Hit, Input, and Output
Pricing: Blended Price
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed: Granite 3.3 8B (Non-reasoning)
Latency vs. Output Speed
Latency
Measured by Time (seconds) to First Token
Time to First 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
| Replicate | 128k | Open | -- | 15 | 27.87 | 61.47 | -- | ||
Frequently Asked Questions
Common questions about Granite 3.3 8B (Non-reasoning) providers
Granite 3.3 8B (Non-reasoning) is available through 1 API provider: Replicate. Each provider offers different performance characteristics and pricing.
Granite 3.3 8B (Non-reasoning) is currently available through 1 API provider that we benchmark and track.
The fastest provider for Granite 3.3 8B (Non-reasoning) by output speed is Replicate (14.6 t/s). Output speed measures how quickly tokens are generated after the model starts responding.
The provider with the lowest time to first token for Granite 3.3 8B (Non-reasoning) is Replicate (28.09s). Lower latency means faster initial response time.
The most affordable provider for Granite 3.3 8B (Non-reasoning) by blended price is Replicate ($0.05 per 1M tokens). Blended price uses a 7:2:1 cache hit/input/output token ratio.
The provider with the lowest input token pricing for Granite 3.3 8B (Non-reasoning) is Replicate ($0.03 per 1M input tokens).
The provider with the lowest output token pricing for Granite 3.3 8B (Non-reasoning) is Replicate ($0.25 per 1M output tokens).
Replicate leads across all key metrics for Granite 3.3 8B (Non-reasoning), offering the fastest speed, lowest latency, and most competitive pricing.
When choosing a provider for Granite 3.3 8B (Non-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 Granite 3.3 8B (Non-reasoning)'s intelligence, capabilities, modalities, and how it compares to other models, see the model overview page. View model overview