Grok 4.20
Grok 4.20 is an API-only (closed weights) language model from xAI, released February 17, 2026. Its makers disclosed about 500B parameters. The context window is 1M tokens. On the Epoch Capabilities Index it scores 152.3, rank 60 of 256 rated models. API price: $1.25 input / $2.5 output per million tokens.
Check it against your own PC →
Lab
xAI
Released
February 17, 2026
Weights
API-only (closed weights)
Parameters
500B disclosed
Context
1M
Price / M tokens
$1.25 / $2.5 in / out
Epoch ECI
152.3 #60
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Math | 37.8 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Knowledge | 69.5 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 92.2% |
| GPQA Diamond | 89.3% |
| Terminal-Bench | 57.3% |
| FrontierMath (Tiers 1–3) | 47.3% |
| SimpleQA Verified | 30.2% |
| Chess Puzzles | 25.3% |
| FrontierMath-Tier-4-v2-Private | 17.1% |
| ProofBench | 14.0% |
Links
Announcement · Epoch AI benchmarks
Models with a similar score
Grok 4.7 · Gemini 3.1 Pro · Qwen3.8 Max (0902) · DeepSeek V4 Pro 0813 · Muse Spark Open · Gemini 3.5 Flash · Gemini 3.6 Flash · Muse Spark 1.2
More from xAI
Grok 4.7 · Grok 4.6 · Grok 4.5 · Grok 4.3 Beta · Grok Build 0.1 · Grok 4.20 Multi-Agent · Grok 4.1 Fast · Grok 4.1