Skip to content
OSMEXO
Sign inGet an API key
☰
Sign inGet an API key

Grok 4.7

xai/grok-4.7
  • xAI
  • Family · Grok 4
  • Released 21 Sep 2026
  • Text · Image · File → Text
  • Reasoning · adjustable
  • Tools · JSON · Vision
Not on OSMEXO yet
Request this model
Spec & list price
Context488K
Max output450K
Input / 1M$2.00
Output / 1M$6.003× input
Scores · standing
Intelligence46.4#17/489
LiveBench70.1#11/32
Usage
Tokens / day36.0B▼ 50% 7d
Usage rank#57
Requests / day419.2K
Tool calls / req0.59
Per request · 30 days
Avg input61.8K4.1× median
Avg output1.4K2.0× median
Cache hit91%1.3× median
Reasoning share76%1.3× median
Tool-call failures0.2%0.1× median

Consumption

66.7B133.3B200.0B3 Oct17 Oct1 Nov
Peak 172.1B 8 OctLow 19.0B 1 NovRank today #57

News & trends

32 mentions this week ▲ 357%this model · videos, posts, articles · 30 days
Trending · 14 days

Benchmarks

bar = share of the top score · Benchmark data: Artificial Analysis
Long-context reasoning#49 of 393
80.0%
SciCode#19 of 160
57.8%
Intelligence index#17 of 489
46.4
Humanity's Last Exam#35 of 465
43.1%
Terminal-Bench 4.0#29 of 152
25.8%
Speed · median across 5 providers, latest day
Output speed#157/33471 tok/sabout median
First token#2421.69 s1.6× longer than median
First answer token#6719 sthinking included

#7 cheapest of 11 with a similar score

Value
Intelligence Index
2030405060$1$2$5$10Grok 4.7
Blended price per 1M tokens (log)
Family progress
Intelligence Index
20304050SepNov2026MarMayJulSepGrok 4.7
  • Family releases46.4
LiveBench suites
Browser49.3
Coding84.5
Data Analysis76.9
Design78.0
Long Agentic46.1
Reasoning82.7
Research86.4

▲ Strongest: Research, Coding. ▼ Weakest: Long Agentic (46).

Technical details

Reasoning
on · high
Tokenizer
Grok
Moderation
none
Reasoning effort
  • xhigh
  • high · default
  • medium
  • low
Request parameters 13
  • tools
  • tool_choice
  • response_format
  • structured_outputs
  • reasoning
  • reasoning_effort
  • include_reasoning
  • temperature
  • top_p
  • max_tokens
+3 more
  • seed
  • logprobs
  • top_logprobs

Interesting facts

  1. 3.5× the output room of similar models (450.0K against a 128.0K median).
  2. The most used of 5 Grok 4 models by tokens last month.
  3. Released 42 days ago, the newest of 6 Grok 4 models.
  4. Weekdays average 71.2B tokens a day against 38.1B on weekends.

When to use

Best for

  • Research and long-form analysis
  • Longer autonomous agent runs
  • Coding assistants with tool calling
  • Image and file understanding

Think twice for

  • Self-hosting (closed weights)
  • Terminal-heavy automation
  • Low-cost high-volume chat

Research papers

Same family

Grok 4

Alternatives

other vendors · similar capability

Want Grok 4.7 on OSMEXO?

Sign in to request it. Requests are reviewed; a request does not guarantee availability.

Request this model on OSMEXO