SpaceXAI: Grok 4.6

Last Verified 2026-10-01 ⚠ sources disagree — shown side by side context 500,000 JSON ↓

⚠ 字段 input 在数据源之间存在分歧——我们不取平均、不抹平，原值并列展示在下方。

Pricing matrix USD per 1M tokens

Field	Consensus	Evidence
Input	Unverified（未验证，不填 0）	disputed ×1.60 3 sources
Output	$6	✓ 3 independent sources 3 sources
Cache read	$0.5	✓ 3 independent sources 3 sources
Cache write (5m)	Unverified（未验证，不填 0）	not published
Cache write (1h)	Unverified（未验证，不填 0）	not published
Reasoning tokens	Unverified（未验证，不填 0）	not published

证据链：各源原值并列（audit trail，点击展开）

Field	Source	Raw value (USD/1M)	Independent	Observed
input	openrouter	2.0000	yes	2026-10-01 08:00
input	litellm	1.2500	yes	2026-10-01 08:00
input	litellm	2.0000	yes	2026-10-01 08:00
input	litellm <span class="badge mute">mirror</span>	2.0000	no	2026-10-01 08:00
input	modelsdev	2.0000	yes	2026-10-01 08:00
output	openrouter	6.0000	yes	2026-10-01 08:00
output	litellm	6.0000	yes	2026-10-01 08:00
output	litellm	6.0000	yes	2026-10-01 08:00
output	litellm <span class="badge mute">mirror</span>	6.0000	no	2026-10-01 08:00
output	modelsdev	6.0000	yes	2026-10-01 08:00
cache_read	openrouter	0.5000	yes	2026-10-01 08:00
cache_read	litellm	0.5000	yes	2026-10-01 08:00
cache_read	litellm	0.5000	yes	2026-10-01 08:00
cache_read	litellm <span class="badge mute">mirror</span>	0.5000	no	2026-10-01 08:00
cache_read	modelsdev	0.5000	yes	2026-10-01 08:00

Long-context tier jumps ⚠

When prompt tokens cross a threshold, the entire request is re-priced at the higher tier — not just the excess. This is the #1 hidden cost on long-context workloads.

Prompt ≥ tokens	input	output	cache read
0 (base)	未验证	$6	$0.5
200,000	未验证	未验证	未验证

Run this model through the real-bill calculator →

Sources

OpenRouter — live API (primary)
LiteLLM — model_prices_and_context_window.json (pinned to commit at fetch time)
models.dev — api.json

This is an estimate based on public sticker prices, not your invoice. Machine-readable: /data/v1/pricing/x-ai-grok-4-6.json