Parasail: Models Intelligence, Performance & Price
Analysis of Parasail's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Parasail for your use-case.
Most Intelligent
GLM-5 (FP8)
Kimi K2.5
Qwen3.5 397B A17BGLM-4.7 (FP8)
MiniMax-M2.5 (FP8)Intelligence index
Total 19 models
Fastest
Llama 4 Maverick (FP8)Qwen3 Next 80B A3B
Olmo 3.1 32B ThinkMolmo2-8BQwen3 Coder Next (FP8)Output speed
Total 19 models
Lowest Price
Olmo 3 7B
gpt-oss-120B (high)gpt-oss-120B (low)Llama 3.3 70B (FP8)
Gemma 3 27BBlended price (per 1M tokens)
Total 19 models
Parasail offers 19 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on Parasail are GLM-5 (FP8) (50), Kimi K2.5 (47), Qwen3.5 397B A17B (45).
- For output speed, the fastest models are Llama 4 Maverick (FP8) (144 t/s), Qwen3 Next 80B A3B (116 t/s), Olmo 3.1 32B Think (107 t/s). Speed varies significantly across models, with a 53% difference between the fastest and slowest.
- For latency, MiniMax-M2.5 (FP8) (0.84s), Qwen3 235B 2507 (1.02s), Qwen3.5 397B A17B (1.03s) offer the lowest time to first token.
- For pricing, Olmo 3 7B ($0.13), gpt-oss-120B (high) ($0.26), gpt-oss-120B (low) ($0.26) offer the lowest blended prices per 1M tokens. Prices vary up to 2.3x across models.
- For context window size, Llama 4 Maverick (FP8) (1m), Qwen3.5 397B A17B (262k), Qwen3 Coder Next (FP8) (262k) support the largest context windows on Parasail.
Intelligence Evaluations
Artificial Analysis Intelligence Index
Intelligence Evaluations
Intelligence vs. Price
Context Window
Context Window
JSON Mode & Function Calling
Function (Tool) Calling & JSON Mode
| Models | Function calling | JSON Mode |
|---|---|---|
| GLM-5 (FP8), Parasail | ||
| Kimi K2.5, Parasail | ||
| Qwen3.5 397B A17B, Parasail | ||
| GLM-4.7 (FP8), Parasail | ||
| MiniMax-M2.5 (FP8), Parasail | ||
| DeepSeek V3.2 (FP8), Parasail | ||
| GLM-4.7 (FP8), Parasail | ||
| gpt-oss-120B (high), Parasail | ||
| Qwen3 Coder Next (FP8), Parasail | ||
| Qwen3 235B 2507, Parasail | ||
| gpt-oss-120B (low), Parasail | ||
| Qwen3 VL 235B A22B (FP8), Parasail |
Pricing
Intelligence vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
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 vs. Price
Key definitions
Frequently Asked Questions
Common questions about Parasail
Parasail offers 19 models that we track: GLM-5, Kimi K2.5, Qwen3.5 397B A17B, GLM-4.7, MiniMax-M2.5, DeepSeek V3.2, GLM-4.7, gpt-oss-120B (high), Qwen3 Coder Next, Qwen3 235B 2507, gpt-oss-120B (low), Qwen3 VL 235B A22B, Qwen3 Next 80B A3B, Llama 4 Maverick, Llama 3.3 70B, Olmo 3.1 32B Think, Gemma 3 27B, Olmo 3 7B, and Molmo2-8B.
The most intelligent model available on Parasail is GLM-5 with an Intelligence Index score of 50.
The fastest model on Parasail by output speed is Llama 4 Maverick at 143.9 tokens per second.
The model with the lowest time to first token on Parasail is MiniMax-M2.5 at 0.84s. Lower latency means faster initial response time.
The most affordable model on Parasail by blended price is Olmo 3 7B at $0.13 per 1M tokens (3:1 input to output ratio).
Yes, Parasail offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
17 of 19 models on Parasail support JSON mode for structured output.
17 of 19 models on Parasail support function calling (tool use).
Yes, Parasail offers 9 reasoning models: GLM-5, Kimi K2.5, Qwen3.5 397B A17B, GLM-4.7, MiniMax-M2.5, DeepSeek V3.2, gpt-oss-120B (high), gpt-oss-120B (low), and Olmo 3.1 32B Think. Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, all 19 models on Parasail are open weight models.
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.
When choosing a model on Parasail, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.