xAI: Grok 4 Fast

x-ai/grok-4-fast

Grok 4 Fast is a xAI model that isn't ranked on the TryAii Score yet — no benchmark results on file. It lists at $0.2/$0.5 per million input/output tokens with a 2M-token context. It's reasonably quick at about 61 tokens/sec.

Prompt $/1M

$0.200

Completion $/1M

$0.500

Context

2,000,000 tok

Best provider

61 tok/s · xAI

Specs

Model id: x-ai/grok-4-fast
Display name: xAI: Grok 4 Fast
Modality: text+image+file->text
Tokenizer: Grok
Instruct type: —
Context length: 2,000,000
Max completion: 30,000
Prompt $/1M: $0.200
Completion $/1M: $0.500
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionyes
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
xAIfastest	61	7,111 ms	9,661 ms

Prompt caching by provider

Provider	Min tokens to cache	Notes
xAI	—	Automatic caching with NO published activation threshold and NO guaranteed TTL (entries may be evicted anytime — the least-documented major platform). Read discount ~75-85% provisional. Cache is per-server: always set the x-grok-conv-id header (or prompt_cache_key on the Responses API) for routing affinity, and keep conversations append-only.