Grok 4.3
Language Available Full comparison ↗ xai/grok-4.3 · by xAI
· decoder-only-transformer
Pricing — 1 offering(s)
About this offering: Requests whose prompt reaches 200k tokens are billed at the higher rate for all tokens in the request. Batch API: 20% off standard prices. Requests to several retired xAI models, including grok-3, now redirect to Grok 4.3 and are billed at its prices.
Input tokens
- $1.25 / 1M tokens (input) Current 2026-05-15 → present
Output tokens
- $2.50 / 1M tokens (output) Current 2026-05-15 → present
Cached input tokens
- $0.20 / 1M tokens (cached input) Current 2026-05-15 → present
Input tokens (prompts >= 200k)
- $2.50 / 1M tokens (input) Current 2026-05-15 → present
Output tokens (prompts >= 200k)
- $5.00 / 1M tokens (output) Current 2026-05-15 → present
Cached input tokens (prompts >= 200k)
- $0.40 / 1M tokens (cached input) Current 2026-05-15 → present
Input tokens (US regional endpoint)
- $1.375 / 1M tokens (input) Current 2026-05-15 → present
Output tokens (US regional endpoint)
- $2.75 / 1M tokens (output) Current 2026-05-15 → present
Showing the active price and any recorded history. Full pricing history is available via the paid API — see API docs.
See how Grok 4.3 fits into a cost-aware routing setup
See how →Capability profile
How Grok 4.3 rates across core capability dimensions, with the task-level evidence behind each rating.
Vendor-reported: xAI's model page describes "strong tool calling", with function calling and structured outputs supported. No independent score recorded here.
Vendor-reported: xAI's model page describes strong "instruction following capabilities". No independent score recorded here.
1,000,000 tokens, per xAI's model page.
Ratings are estimated — limited independent data is available for this model.
Operator guidance
At $1.25 / $2.50 per 1M tokens it costs less than Grok 4.6 and 4.7 ($2 / $6), with twice their context window. It's older than both, so prefer them when capability matters more than price. Reasoning effort defaults to low, and requests with prompts of 200k tokens or more cost double.
Use cases
- Agent and tool-calling workloads where price matters more than xAI's newest capability
- Long-context requests that need more than Grok 4.7's 500K window
Limitations
- Capability ratings are vendor-reported, not independent benchmark runs.
- Prompts of 200k tokens or more are billed at double the rate for all tokens in the request.