MODEL PROVIDER
xAI
xAI’s Grok models through the xAI API, including reasoning, non-reasoning, and coding variants.
xAI has 16 models available in SteelEngine with pricing, context window, and capability details.
MODELS
16
DEFAULT MODEL
Grok 4.6
PRICING
USD PER 1M TOKENS
MODELS
xAI models in SteelEngine.
Recommended models come first. Open any model for its full price sheet, limits, and supported capabilities.
xAI
RECOMMENDED
Grok 4.7
Among the highest-priced models in the xAI lineup, for the hardest reasoning and planning work. A recommended default among SteelEngine’s xAI models. Released Sep 21, 2026.
IN / 1M
$2
OUT / 1M
$6
CONTEXT
500k
xAI
RECOMMENDED
Grok 4.6
Among the highest-priced models in the xAI lineup, for the hardest reasoning and planning work. A recommended default among SteelEngine’s xAI models. Released Aug 12, 2026.
IN / 1M
$2
OUT / 1M
$6
CONTEXT
500k
xAI
Grok 4.1 Fast Reasoning
Mid-priced, latency-focused reasoning model for interactive agent steps. Reads up to 2M tokens of context in one call. Reasons through the request before answering.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
2M
xAI
Grok 4.1 Fast Non Reasoning
Mid-priced, latency-focused model for chat, drafting, and tool calling. Reads up to 2M tokens of context in one call. Answers without a reasoning phase, for lower latency.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
2M
xAI
Grok 4 Fast Reasoning
Mid-priced, latency-focused reasoning model for interactive agent steps. Reads up to 2M tokens of context in one call. Reasons through the request before answering.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
2M
xAI
Grok 4 Fast Non Reasoning
Mid-priced, latency-focused model for chat, drafting, and tool calling. Reads up to 2M tokens of context in one call. Answers without a reasoning phase, for lower latency.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
2M
xAI
Grok 4.3
Mid-priced general model for drafting, summarizing, and tool calling. Reads up to 1M tokens of context in one call. Released Apr 30, 2026.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
1M
xAI
Grok 4.20 0309 Reasoning
Mid-priced reasoning model for everyday agent steps that need deliberate thinking. Reads up to 1M tokens of context in one call.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
1M
xAI
Grok 4.20 0309 Non Reasoning
Mid-priced general model for drafting, summarizing, and tool calling. Reads up to 1M tokens of context in one call. Answers without a reasoning phase, for lower latency.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
1M
xAI
Grok 4.20 Multi Agent 0309
Mid-priced general model for drafting, summarizing, and tool calling. Reads up to 1M tokens of context in one call. Multi-agent variant that coordinates several agents on one request.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
1M
xAI
Grok 4.5
Among the highest-priced models in the xAI lineup, for the hardest generation and analysis work. Handles 500k tokens of context and writes up to 128k tokens.
IN / 1M
$2
OUT / 1M
$6
CONTEXT
500k
xAI
Grok 4 Latest
Mid-priced general model for drafting, summarizing, and tool calling. Handles 256k tokens of context. Alias that tracks the provider’s newest release of this model.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
256k
xAI
Grok 4 0709
Mid-priced general model for drafting, summarizing, and tool calling. Handles 256k tokens of context. Pinned snapshot 0709 for reproducible behavior.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
256k
xAI
Grok Code Fast 1
Lower-priced coding model for completions, small edits, and code-aware tool calls. Handles 256k tokens of context.
IN / 1M
$1
OUT / 1M
$2
CONTEXT
256k
xAI
Grok 3 Latest
Mid-priced general model for drafting, summarizing, and tool calling. Handles 131k tokens of context.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
131k
xAI
Grok 3 Fast Latest
Mid-priced, latency-focused model for chat, drafting, and tool calling. Handles 131k tokens of context.
IN / 1M
$1.25
OUT / 1M
$2.5
CONTEXT
131k
MORE PROVIDERS
Compare other providers.
Every provider page lists the same pricing and capability details, so you can compare like for like.
BUILD WITH STEELENGINE
Put xAI models to work.
Use these models in deterministic workflow steps or production agents, with approvals, guardrails, and a full audit trail around every call.