Grok 4
Proprietary CommercialDeveloped by xAI · Colossus Sparse MoE with Real-Time Reasoning · Modalities: Text, Vision, Real-Time Search · Released 2026-01-14
What are the token costs and operational benchmarks for Grok 4?
Verified Capability Benchmarks
Composite Intelligence: 84.9/100Architectural & Production Profile
xAI's flagship reasoning model trained on the 200k-GPU Colossus cluster, excelling at graduate-level science, code synthesis, and live web telemetry.
Real-time financial/market intelligence, PhD-level STEM problem solving, and autonomous code generation.
- Architecture
- Colossus Sparse MoE with Real-Time Reasoning
- Modalities
- Text, Vision, Real-Time Search
- Max Output Tokens
- 65,536 tokens
- Blended Rate (3:1)
- $5.000 / 1M
Verified Cloud Inference Endpoints Hosting Grok 4
Compare latency, streaming speed, and token pricing across cloud API providers.
| Inference Provider | Role | Input / 1M | Output / 1M | Avg TTFT | Throughput |
|---|---|---|---|---|---|
| xAI Colossus API → | Primary Host | $2.50 | $12.50 | 480ms | 95 tok/s |
| Anysphere Cursor Router → | Partner Endpoint | $2.50 | $12.50 | 495ms | 95 tok/s |
Interactive Monthly Token Economics & ROI Forecaster
Model production workloads across prompt, completion, and cached context tokens, or switch models on the fly.
Estimated Monthly Spend
$2.50/1M in · $12.50/1M out
Prompt Cache: $0.500/1M read · $2.50/1M write
$2.00/1M in · $10.00/1M out
Prompt Cache: $0.100/1M read · $0.50/1M write
Head-to-Head Comparisons Involving Grok 4 (36)
Grok 4 vs Gemini 4 Argon
Gemini 4 Argon is 20.0% cheaper for input tokens ($2.00 vs. $2.50 per 1M tokens) and $10.00 vs. $12.50 for output tokens (1.2x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (370ms faster than Gemini 4 Argon) with 95 vs. 115 tok/s throughput. Gemini 4 Argon leads coding evaluations at 84.5% SWE-bench Verified vs. 73.2% for Grok 4 and 93.8% vs. 86.2% on GPQA Diamond.
Grok 4 vs Claude 4.5 Sonnet
Grok 4 is 16.7% cheaper for input tokens ($2.50 vs. $3.00 per 1M tokens) and $12.50 vs. $15.00 for output tokens (1.2x cost advantage). In terms of operational performance, Claude 4.5 Sonnet delivers lower response latency at 380ms TTFT (100ms faster than Grok 4) with 84 vs. 95 tok/s throughput. Claude 4.5 Sonnet leads coding evaluations at 75.8% SWE-bench Verified vs. 73.2% for Grok 4 and 84.2% vs. 86.2% on GPQA Diamond.
Grok 4 vs Claude Opus 5.5
Grok 4 is 37.5% cheaper for input tokens ($2.50 vs. $4.00 per 1M tokens) and $12.50 vs. $20.00 for output tokens (1.6x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (200ms faster than Claude Opus 5.5) with 95 vs. 95 tok/s throughput. Claude Opus 5.5 leads coding evaluations at 89.9% SWE-bench Verified vs. 73.2% for Grok 4 and 93.2% vs. 86.2% on GPQA Diamond.
Grok 4 vs Gemini 3.8 Flash
Gemini 3.8 Flash is 90.0% cheaper for input tokens ($0.25 vs. $2.50 per 1M tokens) and $1.50 vs. $12.50 for output tokens (10.0x cost advantage). In terms of operational performance, Gemini 3.8 Flash delivers lower response latency at 105ms TTFT (375ms faster than Grok 4) with 210 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 68.5% for Gemini 3.8 Flash and 86.2% vs. 79.2% on GPQA Diamond.
Grok 4 vs GPT-5
GPT-5 and Grok 4 share identical input pricing at $2.50 per 1M tokens ($10.00 vs. $12.50 output). In terms of operational performance, GPT-5 delivers lower response latency at 340ms TTFT (140ms faster than Grok 4) with 105 vs. 95 tok/s throughput. GPT-5 leads coding evaluations at 74.9% SWE-bench Verified vs. 73.2% for Grok 4 and 85.4% vs. 86.2% on GPQA Diamond.
Grok 4 vs GPT-6 Astra
Grok 4 is 75.0% cheaper for input tokens ($2.50 vs. $10.00 per 1M tokens) and $12.50 vs. $50.00 for output tokens (4.0x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (260ms faster than GPT-6 Astra) with 95 vs. 74 tok/s throughput. GPT-6 Astra leads coding evaluations at 75.7% SWE-bench Verified vs. 73.2% for Grok 4 and 96.0% vs. 86.2% on GPQA Diamond.
Grok 4 vs DeepSeek V4 Pro
DeepSeek V4 Pro is 47.2% cheaper for input tokens ($1.32 vs. $2.50 per 1M tokens) and $3.96 vs. $12.50 for output tokens (1.9x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (40ms faster than DeepSeek V4 Pro) with 95 vs. 78 tok/s throughput. DeepSeek V4 Pro leads coding evaluations at 80.6% SWE-bench Verified vs. 73.2% for Grok 4 and 90.1% vs. 86.2% on GPQA Diamond.
Grok 4 vs DeepSeek R1
DeepSeek R1 is 78.0% cheaper for input tokens ($0.55 vs. $2.50 per 1M tokens) and $2.19 vs. $12.50 for output tokens (4.5x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (970ms faster than DeepSeek R1) with 95 vs. 58 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 49.2% for DeepSeek R1 and 86.2% vs. 71.5% on GPQA Diamond.
Grok 4 vs DeepSeek V4 Flash Vision (Exp)
DeepSeek V4 Flash Vision (Exp) is 91.2% cheaper for input tokens ($0.22 vs. $2.50 per 1M tokens) and $0.65 vs. $12.50 for output tokens (11.4x cost advantage). In terms of operational performance, DeepSeek V4 Flash Vision (Exp) delivers lower response latency at 310ms TTFT (170ms faster than Grok 4) with 165 vs. 95 tok/s throughput. DeepSeek V4 Flash Vision (Exp) leads coding evaluations at 77.4% SWE-bench Verified vs. 73.2% for Grok 4 and 91.3% vs. 86.2% on GPQA Diamond.
Grok 4 vs OpenAI o3
OpenAI o3 is 20.0% cheaper for input tokens ($2.00 vs. $2.50 per 1M tokens) and $8.00 vs. $12.50 for output tokens (1.2x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (670ms faster than OpenAI o3) with 95 vs. 82 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 71.7% for OpenAI o3 and 86.2% vs. 87.7% on GPQA Diamond.
Grok 4 vs Claude 4 Opus
Grok 4 is 83.3% cheaper for input tokens ($2.50 vs. $15.00 per 1M tokens) and $12.50 vs. $75.00 for output tokens (6.0x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (440ms faster than Claude 4 Opus) with 95 vs. 46 tok/s throughput. Claude 4 Opus leads coding evaluations at 77.2% SWE-bench Verified vs. 73.2% for Grok 4 and 87.1% vs. 86.2% on GPQA Diamond.
Grok 4 vs Claude 3.7 Sonnet
Grok 4 is 16.7% cheaper for input tokens ($2.50 vs. $3.00 per 1M tokens) and $12.50 vs. $15.00 for output tokens (1.2x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (60ms faster than Claude 3.7 Sonnet) with 95 vs. 72 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 70.3% for Claude 3.7 Sonnet and 86.2% vs. 78.5% on GPQA Diamond.
Grok 4 vs Gemini 3.1 Pro
Gemini 3.1 Pro is 50.0% cheaper for input tokens ($1.25 vs. $2.50 per 1M tokens) and $10.00 vs. $12.50 for output tokens (2.0x cost advantage). In terms of operational performance, Gemini 3.1 Pro delivers lower response latency at 420ms TTFT (60ms faster than Grok 4) with 98 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 71.8% for Gemini 3.1 Pro and 86.2% vs. 82.4% on GPQA Diamond.
Grok 4 vs Llama 4 Maverick
Llama 4 Maverick is 92.0% cheaper for input tokens ($0.20 vs. $2.50 per 1M tokens) and $0.60 vs. $12.50 for output tokens (12.5x cost advantage). In terms of operational performance, Llama 4 Maverick delivers lower response latency at 135ms TTFT (345ms faster than Grok 4) with 175 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 62.4% for Llama 4 Maverick and 86.2% vs. 73.8% on GPQA Diamond.
Grok 4 vs Qwen 3.8 Flash Next
Qwen 3.8 Flash Next is 93.6% cheaper for input tokens ($0.16 vs. $2.50 per 1M tokens) and $0.47 vs. $12.50 for output tokens (15.6x cost advantage). In terms of operational performance, Qwen 3.8 Flash Next delivers lower response latency at 280ms TTFT (200ms faster than Grok 4) with 185 vs. 95 tok/s throughput. Qwen 3.8 Flash Next leads coding evaluations at 81.0% SWE-bench Verified vs. 73.2% for Grok 4 and 91.7% vs. 86.2% on GPQA Diamond.
Grok 4 vs Gemini 3.5 Flash-Lite
Gemini 3.5 Flash-Lite is 97.0% cheaper for input tokens ($0.07 vs. $2.50 per 1M tokens) and $0.30 vs. $12.50 for output tokens (33.3x cost advantage). In terms of operational performance, Gemini 3.5 Flash-Lite delivers lower response latency at 78ms TTFT (402ms faster than Grok 4) with 265 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 55.4% for Gemini 3.5 Flash-Lite and 86.2% vs. 68.4% on GPQA Diamond.
Grok 4 vs OpenAI o4-mini
OpenAI o4-mini is 56.0% cheaper for input tokens ($1.10 vs. $2.50 per 1M tokens) and $4.40 vs. $12.50 for output tokens (2.3x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (200ms faster than OpenAI o4-mini) with 95 vs. 125 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 68.1% for OpenAI o4-mini and 86.2% vs. 81.4% on GPQA Diamond.
Grok 4 vs DeepSeek V3
DeepSeek V3 is 94.4% cheaper for input tokens ($0.14 vs. $2.50 per 1M tokens) and $0.28 vs. $12.50 for output tokens (17.9x cost advantage). In terms of operational performance, DeepSeek V3 delivers lower response latency at 410ms TTFT (70ms faster than Grok 4) with 78 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 42.0% for DeepSeek V3 and 86.2% vs. 59.1% on GPQA Diamond.
Grok 4 vs Gemma 4 31B IT
Gemma 4 31B IT is 95.2% cheaper for input tokens ($0.12 vs. $2.50 per 1M tokens) and $0.36 vs. $12.50 for output tokens (20.8x cost advantage). In terms of operational performance, Gemma 4 31B IT delivers lower response latency at 115ms TTFT (365ms faster than Grok 4) with 155 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 59.2% for Gemma 4 31B IT and 86.2% vs. 71.2% on GPQA Diamond.
Grok 4 vs OpenAI o3-mini
OpenAI o3-mini is 56.0% cheaper for input tokens ($1.10 vs. $2.50 per 1M tokens) and $4.40 vs. $12.50 for output tokens (2.3x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (470ms faster than OpenAI o3-mini) with 95 vs. 98 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 49.3% for OpenAI o3-mini and 86.2% vs. 79.7% on GPQA Diamond.
Grok 4 vs Claude 3.5 Haiku
Claude 3.5 Haiku is 68.0% cheaper for input tokens ($0.80 vs. $2.50 per 1M tokens) and $4.00 vs. $12.50 for output tokens (3.1x cost advantage). In terms of operational performance, Claude 3.5 Haiku delivers lower response latency at 190ms TTFT (290ms faster than Grok 4) with 130 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 40.6% for Claude 3.5 Haiku and 86.2% vs. 41.6% on GPQA Diamond.
Grok 4 vs Claude 3.5 Sonnet
Grok 4 is 16.7% cheaper for input tokens ($2.50 vs. $3.00 per 1M tokens) and $12.50 vs. $15.00 for output tokens (1.2x cost advantage). In terms of operational performance, Claude 3.5 Sonnet delivers lower response latency at 310ms TTFT (170ms faster than Grok 4) with 80 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 49.0% for Claude 3.5 Sonnet and 86.2% vs. 59.4% on GPQA Diamond.
Grok 4 vs Gemini 2.0 Flash
Gemini 2.0 Flash is 96.0% cheaper for input tokens ($0.10 vs. $2.50 per 1M tokens) and $0.40 vs. $12.50 for output tokens (25.0x cost advantage). In terms of operational performance, Gemini 2.0 Flash delivers lower response latency at 280ms TTFT (200ms faster than Grok 4) with 185 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 51.8% for Gemini 2.0 Flash and 86.2% vs. 63.6% on GPQA Diamond.
Grok 4 vs Gemini 2.5 Flash
Gemini 2.5 Flash is 88.0% cheaper for input tokens ($0.30 vs. $2.50 per 1M tokens) and $2.50 vs. $12.50 for output tokens (8.3x cost advantage). In terms of operational performance, Gemini 2.5 Flash delivers lower response latency at 110ms TTFT (370ms faster than Grok 4) with 180 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 58.0% for Gemini 2.5 Flash and 86.2% vs. 70.5% on GPQA Diamond.
Grok 4 vs Gemini 2.5 Pro
Gemini 2.5 Pro is 50.0% cheaper for input tokens ($1.25 vs. $2.50 per 1M tokens) and $10.00 vs. $12.50 for output tokens (2.0x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (220ms faster than Gemini 2.5 Pro) with 95 vs. 80 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 63.8% for Gemini 2.5 Pro and 86.2% vs. 74.0% on GPQA Diamond.
Grok 4 vs GLM-5.3-Flash
GLM-5.3-Flash is 95.2% cheaper for input tokens ($0.12 vs. $2.50 per 1M tokens) and $0.40 vs. $12.50 for output tokens (20.8x cost advantage). In terms of operational performance, GLM-5.3-Flash delivers lower response latency at 90ms TTFT (390ms faster than Grok 4) with 175 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 52.8% for GLM-5.3-Flash and 86.2% vs. 63.5% on GPQA Diamond.
Grok 4 vs GLM-5.3
GLM-5.3 is 86.0% cheaper for input tokens ($0.35 vs. $2.50 per 1M tokens) and $1.00 vs. $12.50 for output tokens (7.1x cost advantage). In terms of operational performance, GLM-5.3 delivers lower response latency at 160ms TTFT (320ms faster than Grok 4) with 120 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 56.0% for GLM-5.3 and 86.2% vs. 67.0% on GPQA Diamond.
Grok 4 vs GPT-4o Mini
GPT-4o Mini is 94.0% cheaper for input tokens ($0.15 vs. $2.50 per 1M tokens) and $0.60 vs. $12.50 for output tokens (16.7x cost advantage). In terms of operational performance, GPT-4o Mini delivers lower response latency at 220ms TTFT (260ms faster than Grok 4) with 140 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 41.0% for GPT-4o Mini and 86.2% vs. 40.2% on GPQA Diamond.
Grok 4 vs GPT-4o
GPT-4o and Grok 4 share identical input pricing at $2.50 per 1M tokens ($10.00 vs. $12.50 output). In terms of operational performance, GPT-4o delivers lower response latency at 280ms TTFT (200ms faster than Grok 4) with 100 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 48.0% for GPT-4o and 86.2% vs. 53.6% on GPQA Diamond.
Grok 4 vs GPT-6.1 Sol
GPT-6.1 Sol is 20.0% cheaper for input tokens ($2.00 vs. $2.50 per 1M tokens) and $10.00 vs. $12.50 for output tokens (1.2x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (40ms faster than GPT-6.1 Sol) with 95 vs. 135 tok/s throughput. GPT-6.1 Sol leads coding evaluations at 86.4% SWE-bench Verified vs. 73.2% for Grok 4 and 92.4% vs. 86.2% on GPQA Diamond.
Grok 4 vs Grok 3 Mini
Grok 3 Mini is 88.0% cheaper for input tokens ($0.30 vs. $2.50 per 1M tokens) and $1.20 vs. $12.50 for output tokens (8.3x cost advantage). In terms of operational performance, Grok 3 Mini delivers lower response latency at 150ms TTFT (330ms faster than Grok 4) with 160 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 54.0% for Grok 3 Mini and 86.2% vs. 68.0% on GPQA Diamond.
Grok 4 vs Grok 3
Grok 4 is 16.7% cheaper for input tokens ($2.50 vs. $3.00 per 1M tokens) and $12.50 vs. $15.00 for output tokens (1.2x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (140ms faster than Grok 3) with 95 vs. 88 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 65.8% for Grok 3 and 86.2% vs. 84.6% on GPQA Diamond.
Grok 4 vs Llama 3.3 70B Instruct
Llama 3.3 70B Instruct is 95.2% cheaper for input tokens ($0.12 vs. $2.50 per 1M tokens) and $0.30 vs. $12.50 for output tokens (20.8x cost advantage). In terms of operational performance, Llama 3.3 70B Instruct delivers lower response latency at 145ms TTFT (335ms faster than Grok 4) with 165 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 42.5% for Llama 3.3 70B Instruct and 86.2% vs. 55.6% on GPQA Diamond.
Grok 4 vs Mistral Large 2 (2411)
Mistral Large 2 (2411) is 20.0% cheaper for input tokens ($2.00 vs. $2.50 per 1M tokens) and $6.00 vs. $12.50 for output tokens (1.2x cost advantage). In terms of operational performance, Mistral Large 2 (2411) delivers lower response latency at 380ms TTFT (100ms faster than Grok 4) with 75 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 45.2% for Mistral Large 2 (2411) and 86.2% vs. 54.2% on GPQA Diamond.
Grok 4 vs o1
Grok 4 is 83.3% cheaper for input tokens ($2.50 vs. $15.00 per 1M tokens) and $12.50 vs. $60.00 for output tokens (6.0x cost advantage). In terms of operational performance, Grok 4 delivers lower response latency at 480ms TTFT (2120ms faster than o1) with 95 vs. 48 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 48.9% for o1 and 86.2% vs. 78.0% on GPQA Diamond.
Grok 4 vs Qwen 3.8 27B
Qwen 3.8 27B is 92.0% cheaper for input tokens ($0.20 vs. $2.50 per 1M tokens) and $0.60 vs. $12.50 for output tokens (12.5x cost advantage). In terms of operational performance, Qwen 3.8 27B delivers lower response latency at 120ms TTFT (360ms faster than Grok 4) with 140 vs. 95 tok/s throughput. Grok 4 leads coding evaluations at 73.2% SWE-bench Verified vs. 58.4% for Qwen 3.8 27B and 86.2% vs. 68.0% on GPQA Diamond.
Frequently Asked Questions & Query Fan-Out
How much does Grok 4 cost per 1M tokens including prompt caching?
Grok 4 costs $2.50 per million input tokens and $12.50 per million output tokens. Cached input reads cost $0.500 per 1M tokens, reducing repeated context costs by 80%.
What are Grok 4's coding and reasoning benchmark scores?
Grok 4 achieves 73.2% on SWE-bench Verified, 95% on HumanEval coding, 86.2% on GPQA Diamond graduate-level science, 89.9% on MMLU-Pro, and 97.6% on MATH-500, earning an overall Inference Value Ratio (IVR) of 57.6/100.
What is the context window, TTFT latency, and token throughput of Grok 4?
Grok 4 supports a 262,144 token context window (65,536 max output tokens) and operates at 480ms average Time to First Token (TTFT) with 95 tokens/sec output streaming throughput.
What are the primary production use cases for Grok 4?
Real-time financial/market intelligence, PhD-level STEM problem solving, and autonomous code generation.