SpaceXAI: Models Intelligence, Performance & Price
This analysis is intended to support you in choosing the best model provided by SpaceXAI for your use-case.
Most Intelligent
UpdatedGrok 4.6 (high)Grok 4.6 (xhigh)Grok 4.6 (medium)Grok 4.5 (high)Grok 4.6 (low)Intelligence index
Total 15 models
Fastest
Grok 4.3 (high)Grok 4.20 0309 v2Grok 4.3 (medium)Grok 4.3 (Non-reasoning)Grok 3 FastOutput speed
Total 15 models
Lowest Price
Grok 3 mini Reasoning (high)Grok Build 0.1 0616Grok 3 mini Reasoning (high) FastGrok 4.20 0309 v2Grok 4.3 (high)Blended price (per 1M tokens)
Total 15 models
SpaceXAI offers 15 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 SpaceXAI are Grok 4.6 (high) (44), Grok 4.6 (xhigh) (44), and Grok 4.6 (medium) (43).
- For output speed, the fastest models are Grok 4.3 (high) (119 t/s), Grok 4.20 0309 v2 (113 t/s), and Grok 4.3 (medium) (108 t/s).
- For latency, Grok 4.20 0309 v2 (Non-reasoning) (0.59s), Grok 3 Fast (0.72s), and Grok 4.3 (Non-reasoning) (0.83s) offer the lowest time to first answer token.
- For pricing, Grok 3 mini Reasoning (high) ($0.16), Grok Build 0.1 0616 ($0.54), and Grok 3 mini Reasoning (high) Fast ($0.57) offer the lowest blended prices per 1M tokens. Prices vary up to 3.9x across models.
- For context window size, Grok 4.20 0309 v2 (2M), Grok 4.20 0309 v2 (Non-reasoning) (2M), and Grok 4.3 (high) (1M) support the largest context windows on SpaceXAI.
Highlights
Intelligence Evaluations
Artificial Analysis Intelligence Index
Intelligence Evaluations
Agentic knowledge work, (Elo-500)/2000
Agentic real-world work tasks, (Elo-500)/2000
Agentic SaaS workflows
Agentic coding & terminal use
Coding
Reasoning & knowledge
Professional document reasoning, All-pass
Physics reasoning
Knowledge
1 - hallucination rate
Long context reasoning
Legal agentic work, criterion pass rate
Agentic business operations
Quantitative analysis on spreadsheets & documents
Agentic tool use
Kubernetes incident root-cause analysis
Visual reasoning
Medical long context reasoning
Intelligence Index vs. Price
Context Window
Context Window
Pricing
Intelligence Index 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
Latency: 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
Further Analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Grok 4.6 (high) | 500k | Proprietary | 44 | $1.86 | 63 | 38.63 | 46.59 | -- | |||
Grok 4.6 (xhigh) | 500k | Proprietary | 44 | $2.32 | 57 | 25.40 | 34.11 | -- | |||
Grok 4.6 (medium) | 500k | Proprietary | 43 | $1.50 | 56 | 31.59 | 40.52 | -- | |||
Grok 4.5 (high) | 500k | Proprietary | 39 | $1.04 | 60 | 7.02 | 15.29 | -- | |||
Grok 4.6 (low) | 500k | Proprietary | 35 | $0.48 | 54 | 5.90 | 15.14 | -- | |||
Grok Build 0.1 0616 | 256k | Proprietary | 27* | -- | 51 | 0.59 | 49.97 | 39.50 | |||
Grok 4.20 0309 v2 | 2M | Proprietary | 26* | -- | 113 | 16.02 | 20.42 | -- | |||
Grok 4.3 (high) | 1M | Proprietary | 25 | $0.17 | 119 | 19.61 | 23.82 | -- | |||
Grok 4.20 0309 | 2M | Proprietary | 25* | -- | -- | -- | -- | -- | |||
Grok 4.3 (medium) | 1M | Proprietary | 25* | -- | 108 | 12.57 | 17.20 | -- | |||
Grok 4.3 (low) | 1M | Proprietary | 24* | -- | 102 | 4.55 | 9.43 | -- | |||
Grok 3 mini Reasoning (high) | 131k | Proprietary | 15* | -- | 75 | 0.68 | 33.97 | 26.63 | |||
Grok 3 mini Reasoning (high) Fast | 131k | Proprietary | 15* | -- | 80 | 0.70 | 31.89 | 24.95 | |||
Grok 4.20 0309 (Non-reasoning) | 2M | Proprietary | 15* | -- | -- | -- | -- | -- | |||
Grok 4.3 (Non-reasoning) | 1M | Proprietary | 15 | $0.14 | 107 | 0.83 | 5.50 | -- | |||
Grok 4.20 0309 v2 (Non-reasoning) | 2M | Proprietary | 14* | -- | 100 | 0.59 | 5.60 | -- | |||
Grok 3 Fast | 131k | Proprietary | 12* | -- | 103 | 0.72 | 5.55 | -- | |||
Key definitions
Frequently Asked Questions
Common questions about SpaceXAI
SpaceXAI offers 15 models that we track: Grok 4.6 (high), Grok 4.6 (xhigh), Grok 4.6 (medium), Grok 4.5 (high), Grok 4.6 (low), Grok Build 0.1 0616, Grok 4.20 0309 v2, Grok 4.3 (high), Grok 4.3 (medium), Grok 4.3 (low), Grok 3 mini Reasoning (high), Grok 3 mini Reasoning (high) Fast, Grok 4.3 (Non-reasoning), Grok 4.20 0309 v2 (Non-reasoning), and Grok 3 Fast.
The most intelligent model available on SpaceXAI is Grok 4.6 (high) with an Intelligence Index score of 44.
The fastest model on SpaceXAI by output speed is Grok 4.3 (high) at 119.0 tokens per second.
The model with the lowest time to first answer token on SpaceXAI is Grok 4.20 0309 v2 (Non-reasoning) at 0.59s. Lower latency means faster initial response time.
The most affordable model on SpaceXAI by blended price is Grok 3 mini Reasoning (high) at $0.16 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on SpaceXAI vary up to 22x across models, from $0.16 per 1M tokens for Grok 3 mini Reasoning (high) to $3.55 per 1M tokens for Grok 3 Fast.
Yes, SpaceXAI offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 15 models on SpaceXAI support JSON mode for structured output.
Yes, all 15 models on SpaceXAI support function calling (tool use).
Yes, SpaceXAI offers 12 reasoning models: Grok 4.6 (high), Grok 4.6 (xhigh), Grok 4.6 (medium), Grok 4.5 (high), Grok 4.6 (low), Grok Build 0.1 0616, Grok 4.20 0309 v2, Grok 4.3 (high), Grok 4.3 (medium), Grok 4.3 (low), Grok 3 mini Reasoning (high), and Grok 3 mini Reasoning (high) Fast. Reasoning models use extended thinking to work through complex problems before providing an answer.
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 SpaceXAI, 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.