Grok 4.20 (Reasoning)
No shutdown before this date. Grok 4.20 (Reasoning) stays on Azure AI Foundry until at least Apr 6, 2027.
≥ 6 mo- Earliest retirement
- ≥ 174 daysApr 6, 2027 · Azure AI Foundry
- GPQA Diamond
- 89.3%#42 of 108 · Epoch AI
- FrontierMath Tier 4
- 17.1%#44 of 58 · Epoch AI
- List price · $ per 1M
- $1.25 in $2.50 outxAI API · cached input $0.20
- Context window
- 1Mtokens
Dates on every platform
2 listings · a floor is the earliest possible date
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Azure AI FoundryMicrosoft: Model retirement schedule - Microsoft Foundry (opens in a new tab) · checked Oct 14, 2026 | grok-4-20-reasoning:1Preview | – | ≥ 6 moNot sooner than | – |
| xAI APIxAI: grok-4.20-0309-reasoning model page (opens in a new tab) · checked Oct 14, 2026us-east-1, us-west-2 | grok-4.20-0309-reasoningActive · also grok-4.20-reasoning-latest, grok-4.20, grok-4.20-reasoning, grok-4.20-0309, grok-4.20-beta-0309-reasoning, grok-4.20-beta, grok-4.20-beta-0309, grok-4.20-beta-latest, grok-4.20-beta-latest-reasoning, grok-4.20-beta-reasoning, grok-4.20-experimental-beta-0304-reasoning, grok-4.20-experimental-beta-0304, grok-4.20-experimental-beta-reasoning-latest, grok-4.20-experimental-beta-latest, grok-4.20-reasoning-gv2 | – | None set | – |
Preview dates on Azure are a floor. Preview deployments are upgraded to a newer version or retired with at least 30 days' notice.
logprobs and top_logprobs are silently ignored by grok-4.20 and newer models.
What breaks when you move off Grok 4.20 (Reasoning)
No breaking changes are recorded for Grok 4.20 (Reasoning).
Intelligence
Third-party results, credited below
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| GPQA Diamond | Epoch AI | 89.3% ± 1.9% | #42 of 108 | Jul 13, 2026 |
| FrontierMath Tier 4 | Epoch AI | 17.1% ± 5.9% | #44 of 58 | Jul 13, 2026 |
| FrontierMath Tiers 1-3 | Epoch AI | 44.9% ± 3.0% | #49 of 71 | Jul 13, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 92.2% ± 3.3% | #42 of 105 | Jul 13, 2026 |
| SimpleQA Verified | Epoch AI | 30.2% ± 1.5% | #55 of 68 | Aug 27, 2026 |
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
$ per 1M tokens, with each price’s source
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
xAI APIgrok-4.20-0309-reasoning | $1.25 | $0.20 | $2.50 | Oct 14, 2026Current · Requests whose prompt reaches 200k tokens are billed at twice these rates for all tokens in the request. Batch API requests are billed at a 20% discount to standard rates. | xAI: Models (opens in a new tab) |