Grok 4.3
No shutdown before this date. Grok 4.3 stays on Amazon Bedrock until at least Jun 15, 2027.
≥ 8 mo- Earliest retirement
- ≥ 244 daysJun 15, 2027 · Amazon Bedrock
- Arena Text
- 1,442#49 of 128 · Arena
- GPQA Diamond
- 88.8%#43 of 108 · Epoch AI
- List price · $ per 1M
- $1.25 in $2.50 outxAI API · cached input $0.20
- Context window
- 1Mtokens
Dates on every platform
| Platform | Model ID and status | Deprecated | Shuts down | Replacement |
|---|---|---|---|---|
| Amazon BedrockAmazon Web Services: Amazon Bedrock model card: Grok 4.3 (opens in a new tab) · checked Oct 14, 2026 | xai.grok-4.3Active | – | ≥ 8 moNot sooner than | – |
| xAI APIxAI: grok-4.3 model page (opens in a new tab) · checked Oct 14, 2026us-east-1, eu-west-1, us-west-2 | grok-4.3Active · also grok-4.3-latest | – | None set | – |
Served on the bedrock-mantle endpoint only.
Not measured on this platform under its terms.
Default reasoning effort is low; supported efforts are none, low, medium, high and xhigh.
Since 15 May 2026 it also serves requests sent to the retired grok-4-1-fast, grok-4-fast, grok-4-0709 and grok-3 slugs.
What breaks when you move off Grok 4.3
No breaking changes are recorded for Grok 4.3.
Intelligence
| Benchmark | Publisher | Score | Rank | Published |
|---|---|---|---|---|
| Arena Text | Arena | 1,442 ± 2 | #49 of 128 | Oct 8, 2026 |
| GPQA Diamond | Epoch AI | 88.8% ± 2.0% | #43 of 108 | Jun 17, 2026 |
| Arena WebDev | Arena | 1,357 ± 3 | #53 of 69 | Oct 8, 2026 |
| FrontierMath Tier 4 | Epoch AI | 14.6% ± 5.6% | #46 of 58 | Jun 17, 2026 |
| FrontierMath Tiers 1-3 | Epoch AI | 42.8% ± 2.9% | #50 of 71 | Jun 17, 2026 |
| OTIS Mock AIME 2024-2025 | Epoch AI | 93.3% ± 3.0% | #39 of 105 | Jun 17, 2026 |
| SimpleQA Verified | Epoch AI | 33.2% ± 1.5% | #51 of 68 | Aug 27, 2026 |
Arena (arena.ai), https://arena.ai/leaderboard
Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource].
List prices
| Platform | Input | Cached | Output | In effect | Source |
|---|---|---|---|---|---|
Amazon Bedrockxai.grok-4.3 | $1.25 | $0.20 | $2.50 | Oct 1, 2026Current · Standard tier on the bedrock-mantle endpoint. | Amazon Web Services: AWS Price List API: AmazonBedrock, us-east-1 (opens in a new tab) |
xAI APIgrok-4.3 | $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) |
Change history
RedirectSix Grok models retired on the xAI API and redirected to Grok 4.3xAI API3 breaks
xAI retired grok-4-1-fast-reasoning, grok-4-1-fast-non-reasoning, grok-4-fast-reasoning, grok-4-fast-non-reasoning, grok-4-0709 and grok-3. The slugs still resolve: requests are served by grok-4.3 instead of returning an error.
- Platform
- xAI API
- Models
- Grok 4.1 Fast (Reasoning), Grok 4.1 Fast (Non-Reasoning), Grok 4 Fast (Reasoning), Grok 4 Fast (Non-Reasoning), Grok 4, Grok 3
- Replacement
- Grok 4.3
- PricingRequests to the retired slugs are billed at grok-4.3 rates, not the rates of the original model.
- DefaultsReasoning slugs are served with low reasoning effort and non-reasoning slugs with none; set medium or high explicitly for deeper reasoning.
- Parametersgrok-4.20 and newer models silently ignore logprobs and top_logprobs.
Source: xAI, Grok Model Retirement on May 15, 2026 (opens in a new tab); xAI, Models (opens in a new tab)