Model Comparison
Grok 4.20 0309 (Reasoning)
vs. Grok 4.3 (Non-reasoning)
Comparing 2 AI models · 7 benchmarks · xAI
Recommended Pick
xA Grok 4.20 0309 (Reasoning) 11 metric wins Strongest on: Throughput, Reasoning, Intelligence
Best Value
Grok 4.3 (Non-reasoning)
100.0 value score
34.4 reasoning / $1.56/1M
Lowest Price
Grok 4.3 (Non-reasoning)
$1.25/1M input price
Best Reasoning
Grok 4.20 0309 (Reasoning)
55.7 reasoning score
Blends available reasoning benchmarks
Best for Coding
Grok 4.20 0309 (Reasoning)
42.2 coding index
Composite Indices
Higher is better; speed and price are normalized
Standard Benchmarks
Only benchmarks with data are shown
Differences That Matter
Best value
Grok 4.3 (Non-reasoning) has the strongest quality-to-price mix at 100.0 out of 100 value points.
Price gap
Grok 4.3 (Non-reasoning) is 1.6x cheaper on input tokens than Grok 4.20 0309 (Reasoning).
Speed gap
Grok 4.20 0309 (Reasoning) generates about 1.3x as many tokens per second as Grok 4.3 (Non-reasoning).
Reasoning gap
Grok 4.20 0309 (Reasoning) leads Grok 4.3 (Non-reasoning) by 21.2 points on reasoning.
Coding gap
Grok 4.20 0309 (Reasoning) leads Grok 4.3 (Non-reasoning) by 17.1 points on coding.
Response Face-Off
Run one prompt through the selected models and compare response quality with live speed and cost context.
Grok 4.20 0309 (Reasoning)
xAI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Grok 4.3 (Non-reasoning)
xAI
TTFT
—
Time
—
tok/s
—
Tokens
—
Cost
—
Which answer was more useful?
Full Comparison
| Metric | Top Pick xA Grok 4.20 0309 (Reasoning) | xA Grok 4.3 (Non-reasoning) |
|---|---|---|
| Pricing per 1M tokens | ||
| Input Cost | $2.00/1M | $1.25/1M |
| Output Cost | $6.00/1M | $2.50/1M |
| Blended (3:1) | $3.00/1M | $1.56/1M |
| Specifications | ||
| Organization | xAI | xAI |
| Release Date | Mar 10, 2026 | Apr 30, 2026 |
| Performance & Speed | ||
| Throughput | 175.2 tok/s | 135.6 tok/s |
| TTFT | 13254ms | 560ms |
| Latency | 13254ms | 560ms |
| Composite Indices | ||
| Value Score | 84.2 | 100.0 |
| Reasoning Score | 55.7 | 34.4 |
| Intelligence | 48.5 | 31.0 |
| Coding | 42.2 | 25.1 |
| Standard Benchmarks | ||
| GPQA | 88.5% | 65.8% |
| HLE | 30.0% | 6.5% |
| SciCode | 44.7% | 37.4% |
| LCR | 59.0% | 24.7% |
| IFBench | 82.9% | 47.6% |
| TAU-bench v2 | 96.5% | 65.8% |
| TerminalBench Hard | 40.9% | 18.9% |
Key Takeaways
Grok 4.3 (Non-reasoning) offers the best value at $1.25/1M, making it ideal for high-volume applications and cost-conscious projects.
Grok 4.20 0309 (Reasoning) has the strongest reasoning profile with a 55.7 reasoning score, combining the available reasoning-heavy benchmarks.
Grok 4.20 0309 (Reasoning) reaches a 42.2 coding index, making it the top choice for software development and code generation tasks.
All models support context windows of ∞+ tokens, suitable for processing lengthy documents and maintaining extended conversations.
When to Choose Each Model
Grok 4.20 0309 (Reasoning)
- Complex reasoning tasks
- Research & analysis
- Code generation
- Software development
Grok 4.3 (Non-reasoning)
- Cost-sensitive applications
- High-volume processing
