Databricks: Models Intelligence, Performance & Price
Analysis of Databricks'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 Databricks for your use-case.
(te)
Most Intelligent
GPT-5.2 (xhigh)
Gemini 3 Pro Preview (high)GPT-5.1 (high)GPT-5 (high)
Claude Opus 4.5Intelligence index
Total 20 models
Fastest
gpt-oss-20B (low)gpt-oss-20B (high)
Llama 3.1 8Bgpt-oss-120B (high)gpt-oss-120B (low)Output speed
Total 20 models
Lowest Price
gpt-oss-20B (high)gpt-oss-20B (low)GPT-5 nano (high)Llama 3.1 8BGemma 3 12BBlended price (per 1M tokens)
Total 20 models
Databricks offers 20 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 Databricks are GPT-5.2 (xhigh) (51), Gemini 3 Pro Preview (high) (48), GPT-5.1 (high) (48).
- For output speed, the fastest models are gpt-oss-20B (low) (302 t/s), gpt-oss-20B (high) (300 t/s), Llama 3.1 8B (167 t/s). Speed varies significantly across models, with a 92% difference between the fastest and slowest.
- For latency, gpt-oss-120B (high) (0.61s), gpt-oss-20B (low) (0.61s), gpt-oss-120B (low) (0.61s) offer the lowest time to first token.
- For pricing, gpt-oss-20B (high) ($0.13), gpt-oss-20B (low) ($0.13), GPT-5 nano (high) ($0.14) offer the lowest blended prices per 1M tokens.
- For context window size, Gemini 3 Pro Preview (high) (1m), Claude 4.5 Sonnet (1m), Claude 4 Sonnet (1m) support the largest context windows on Databricks.
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 |
|---|---|---|
 GPT-5.2 (xhigh), Databricks | ||
| Gemini 3 Pro Preview (high), Databricks | ||
| GPT-5.1 (high), Databricks | ||
| GPT-5 (high), Databricks | ||
| Claude Opus 4.5, Databricks | ||
| GPT-5 mini (high), Databricks | ||
| Claude 4.5 Sonnet, Databricks | ||
| gpt-oss-120B (high), Databricks | ||
| Claude 4 Sonnet, Databricks | ||
| Claude 3.7 Sonnet, Databricks | ||
| GPT-5 nano (high), Databricks | ||
| gpt-oss-120B (low), Databricks |
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 Databricks
Databricks offers 20 models that we track: GPT-5.2 (xhigh), Gemini 3 Pro Preview (high), GPT-5.1 (high), GPT-5 (high), Claude Opus 4.5, GPT-5 mini (high), Claude 4.5 Sonnet, gpt-oss-120B (high), Claude 4 Sonnet, Claude 3.7 Sonnet, GPT-5 nano (high), gpt-oss-120B (low), gpt-oss-20B (high), gpt-oss-20B (low), Llama 4 Maverick, Llama 3.1 405B, Llama 3.3 70B, Llama 3.1 8B, Gemma 3 12B, and Claude 4.1 Opus.
The most intelligent model available on Databricks is GPT-5.2 (xhigh) with an Intelligence Index score of 51.
The fastest model on Databricks by output speed is gpt-oss-20B (low) at 301.8 tokens per second.
The model with the lowest time to first token on Databricks is gpt-oss-120B (high) at 0.61s. Lower latency means faster initial response time.
The most affordable model on Databricks by blended price is gpt-oss-20B (high) at $0.13 per 1M tokens (3:1 input to output ratio).
Prices on Databricks vary up to 235x across models, from $0.13 per 1M tokens for gpt-oss-20B (high) to $30.00 per 1M tokens for Claude 4.1 Opus.
Yes, Databricks offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
7 of 20 models on Databricks support JSON mode for structured output.
16 of 20 models on Databricks support function calling (tool use).
Yes, Databricks offers 10 reasoning models: GPT-5.2 (xhigh), Gemini 3 Pro Preview (high), GPT-5.1 (high), GPT-5 (high), GPT-5 mini (high), gpt-oss-120B (high), GPT-5 nano (high), gpt-oss-120B (low), gpt-oss-20B (high), and gpt-oss-20B (low). Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, 9 of 20 models on Databricks are open weight models: gpt-oss-120B (high), gpt-oss-120B (low), gpt-oss-20B (high), gpt-oss-20B (low), Llama 4 Maverick, Llama 3.1 405B, Llama 3.3 70B, Llama 3.1 8B, and Gemma 3 12B.
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 Databricks, 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.