Grok 4 vs Grok 4-20
Complete pricing and performance comparison between xAI's Grok 4 and xAI's Grok 4-20.
Quick Verdict
Cheaper
Grok 4-20
1.5x cheaper input, 2.5x cheaper output
Larger Context
Grok 4
2.0M vs 131K
Higher Quality
Grok 4-20
Score: 49 vs 42
Faster
Grok 4-20
101 vs 43 tok/s
Pricing Comparison
| Spec | Grok 4 | Grok 4-20 | Difference |
|---|---|---|---|
| Provider | xAI | xAI | |
| Input / 1M tokens | $3 | $2 | Grok 4-20 is 33% more expensive |
| Output / 1M tokens | $15 | $6 | Grok 4-20 is 60% more expensive |
| Context Window | 2.0M | 131K | 15x difference |
| Max Output | 16K | 16K |
Performance Benchmarks
| Metric | Grok 4 | Grok 4-20 | Winner |
|---|---|---|---|
| Quality Index | 42 | 49 | Grok 4-20 |
| Output Speed | 43 tok/s | 101 tok/s | Grok 4-20 |
| Time to First Token | 8.08s | 20.69s | Grok 4 |
| Value (Quality/$) | 13.8 | 24.3 | Higher = better value |
Benchmark data from Artificial Analysis. Quality Index is a composite score across reasoning, coding, and knowledge tasks.
Cost at Scale
Estimated cost at different usage levels (3:1 input-to-output token ratio, typical for chat).
| Usage | Grok 4 | Grok 4-20 | Savings |
|---|---|---|---|
Single request 1K in / 300 out | $0.0075 | $0.0038 | Grok 4-20 saves $0.0037 |
10 requests 10K in / 3K out | $0.075 | $0.038 | Grok 4-20 saves $0.037 |
100 requests 100K in / 30K out | $0.750 | $0.380 | Grok 4-20 saves $0.370 |
1,000 requests 1M in / 300K out | $7.50 | $3.80 | Grok 4-20 saves $3.70 |
10,000 requests 10M in / 3M out | $75.00 | $38.00 | Grok 4-20 saves $37.00 |
1M requests/mo 1B in / 300M out | $7500.00 | $3800.00 | Grok 4-20 saves $3700.00 |
Pros & Cons
Grok 4 Strengths
- +Larger context window (2.0M vs 131K)
- +Lower latency (faster first token)
Grok 4-20 Strengths
- +Cheaper input tokens
- +Cheaper output tokens
- +Faster output (101 vs 43 tok/s)
- +Higher quality score (49 vs 42)
When to Use Each Model
Choose Grok 4 for
- →Long documents, large codebases, or multi-turn conversations
Choose Grok 4-20 for
- →Budget-conscious projects where cost is the primary factor
- →Tasks requiring maximum accuracy and reasoning
- →Real-time applications, chat, or autocomplete
Frequently Asked Questions
Which is cheaper, Grok 4 or Grok 4-20?
For input tokens, Grok 4-20 is 1.5x cheaper at $2/1M tokens. For output tokens, Grok 4-20 is 2.5x cheaper at $6/1M tokens. At typical usage (1M input + 300K output), Grok 4 costs $7.50 vs Grok 4-20 at $3.80.
What's the context window difference?
Grok 4 supports 2.0M context (2,000,000 tokens), while Grok 4-20 supports 131K (131,072 tokens). Grok 4 can handle 15x more context in a single request.
Which model has better benchmarks?
Quality Index: Grok 4 scores 42 vs Grok 4-20 at 49. Speed: Grok 4 generates 43 tok/s vs Grok 4-20 at 101 tok/s. Time to first token: Grok 4 at 8.08s vs Grok 4-20 at 20.69s.
When should I choose Grok 4 over Grok 4-20?
Choose Grok 4 when you need: Larger context window (2.0M vs 131K), Lower latency (faster first token). Choose Grok 4-20 when you need: Cheaper input tokens, Cheaper output tokens, Faster output (101 vs 43 tok/s), Higher quality score (49 vs 42).
How much would 10,000 API requests cost?
At 1K input + 300 output tokens per request (typical chat): Grok 4 = $75.00, Grok 4-20 = $38.00. At 10K input + 1K output per request (longer conversations): Grok 4 = $450.00, Grok 4-20 = $260.00.