xAI Grok 4.3 API Pricing: Real-Time Intelligence & Streaming Speed
Examine xAI Grok 4.3 API pricing ($1.25/M input, $2.50/M output), 1M context token costs, real-time X search integration, and high-speed streaming.
Full specs, context window and API limits →How much does Grok 4.3 cost per million tokens?
xAI Grok 4.3 charges $1.25 per million input tokens and $2.50 per million output tokens ($1.5625/M blended at 3:1). Features 1M context window, high-speed streaming, and live X real-time grounding. Verified 2026-09-08.
How much does Grok 4.3 cost per 1,000 requests?
Computed from generated token pricing. Each row assumes the listed input and output tokens per request; output is adjusted by this model's measured 1.41× verbosity factor.
| Request shape | Input tokens | Output tokens | Cost / 1,000 requests |
|---|---|---|---|
| Short | 100 | 50 | $0.3012 |
| Medium | 1,000 | 500 | $3.0125 |
| Long | 4,000 | 2,000 | $12.0500 |
Formula: ((input price × input tokens) + (output price × output tokens × verbosity factor)) ÷ 1,000,000 × 1,000. Assumptions: short 100/50, medium 1,000/500, long 4,000/2,000 input/output tokens per request. Verbosity run: 2026-06-16T20:31:30.728Z.
Three model-specific pricing decisions
Grok 4.3 owns its lower-rate exact-model bill. The 4.20 reasoning route is a calculation input only, and its output expansion is visible rather than hidden.
1. Chat, long-document, and agent monthly bills
| Workload | Fixed input / output | 100K requests | Unit note |
|---|---|---|---|
| Chat | 1,000 / 300 | $200.00 | Text tokens |
| Long document | 32,000 / 2,000 | $4500.00 | Text tokens |
| Agent | 8,000 / 1,200 | $1300.00 | Retry rate not assumed |
2. Required uplift for Grok 4.20 Reasoning
| Fixed input / output | Grok 4.3 | Grok-4.20 Reasoning | Narrow decision boundary |
|---|---|---|---|
| Coding reasoning · 4,000 / 1,000 | $750.00 | $1400.00 | 15% accepted-result uplift required |
| Long agent · 16,000 / 2,000 | $2500.00 | $4400.00 | 20% accepted-result uplift required |
Formula: requests × (input tokens × input $/M + output tokens × output $/M) ÷ 1,000,000. The uplift is a planning threshold, not a measured quality claim.
3. Price / latency / retry frontier and evidence matrix
| Dimension | Dated evidence | Safe treatment |
|---|---|---|
| Price | $1.25 input / $2.50 output per M | Sourced registry rate |
| Latency | 98 tokens/sec; TTFT 320 ms; 5 measured samples | Measured/estimated state stays visible |
| Retry | Unavailable | Adjusted cost = base ÷ (1 − retry rate) |
| Context / cache / batch / rate limits / SLA | Unavailable | No zero or free default |
Verified 2026-05-19. Luna is the data owner. “Unavailable” means no compatible dated evidence was found; it is never treated as zero. First-party price source · Run this Batch 5 scenario.
All three Batch 5 decisions are server-rendered for Grok 4.3; fixed inputs, formulas, dated sources, speed sample state, and unavailable mechanics are visible.
Batch 68 · exact model pricing decision contributions · verified 2026-09-08
Exact model boundary: xAI grok-4.3 (slug grok-4-3). First-party provider pricing and API documentation remain fact owners.
Frontier speed token pricing and monthly spend matrix
Frozen Batch 68 scenario board. Formula / deterministic rule: monthly_spend = calls * ((in_tokens * 1.25 + out_tokens * 2.50) / 1M) Boundary: Owns Grok 4.3 standard generation token tariffs and monthly volume modeling.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch68-grok-4-3-m1-r125K real-time breaking news queries | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=25K real-time breaking news queries; workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 25K real-time breaking news queries is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m1-r2100K customer interaction sessions | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=100K customer interaction sessions; workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 100K customer interaction sessions is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m1-r3500K automated live-data monitoring passes | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=500K automated live-data monitoring passes; workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 500K automated live-data monitoring passes is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m1-r4prompt caching reuse (50% input discount) | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=prompt caching reuse (50% input discount); workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — prompt caching reuse (50% input discount) is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m1-r5batch processing API queue | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=batch processing API queue; workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — batch processing API queue is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m1-r6unresolved billing currency | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=unresolved billing currency; workload; prompt tokens; completion tokens; standard spend; cached spend; cost winner; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — unresolved billing currency has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: xAI Grok developer pricing; verification date 2026-09-08. Missing or conflicting joins fail closed.
Live X search retrieval grounding and latency audit
Frozen Batch 68 scenario board. Formula / deterministic rule: total_latency = search_fetch_time + ttft + (tokens_out / tps) Boundary: Owns live data search integration economics and retrieval overhead.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch68-grok-4-3-m2-r1real-time social trend analysis | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=real-time social trend analysis; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — real-time social trend analysis is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m2-r2financial market breaking event query | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=financial market breaking event query; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — financial market breaking event query is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m2-r3historical public sentiment research | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=historical public sentiment research; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — historical public sentiment research is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m2-r4high-concurrency live monitoring loop | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=high-concurrency live monitoring loop; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — high-concurrency live monitoring loop is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m2-r5rate-limited search provider fallback | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=rate-limited search provider fallback; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — rate-limited search provider fallback is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m2-r6unsupported query syntax | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=unsupported query syntax; query type; live search units; grounding latency; output tokens; retrieval accuracy; grounding verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — unsupported query syntax has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: xAI Grok documentation; verification date 2026-09-08. Missing or conflicting joins fail closed.
Context window scaling and 1M token cache break-even
Frozen Batch 68 scenario board. Formula / deterministic rule: cache_cost = (tokens * 0.625) / 1M + storage_fee; un-cached = (tokens * 1.25) / 1M Boundary: Owns long-context economics and prompt cache amortization for Grok 4.3.
| Frozen scenario / field ID | Model, identity, provider, and evidence fields | Result | State |
|---|---|---|---|
batch68-grok-4-3-m3-r1128K document context pass | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=128K document context pass; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 128K document context pass is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m3-r2500K codebase exploration session | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=500K codebase exploration session; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 500K codebase exploration session is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m3-r31M maximum context window analysis | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=1M maximum context window analysis; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — 1M maximum context window analysis is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m3-r4multi-turn live monitoring workspace | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=multi-turn live monitoring workspace; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — multi-turn live monitoring workspace is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m3-r5cache eviction before 5-minute TTL | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=cache eviction before 5-minute TTL; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — cache eviction before 5-minute TTL is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
batch68-grok-4-3-m3-r6unresolved cache storage retention | model=grok-4.3; slug=grok-4-3; provider=xAI; scenario=unresolved cache storage retention; context depth; base input cost; cached input cost; minimum reuse break-even; caching recommendation; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-08; measurement versus assumption=explicit | Unavailable — unresolved cache storage retention has no matched, dated bilateral observation. | FAIL CLOSED — manual, probe, or source evidence required |
First-party provenance: xAI Grok developer pricing; verification date 2026-09-08. Missing or conflicting joins fail closed.
Method and limitations: formulas are deterministic; observed and assumed inputs are labeled; no missing provider, host, account, region, realm, alias, snapshot, revision, weight, artifact, control, tool, modality, workload, rate-period, timestamp, or result is transferred. Run the grok-4-3 Batch 68 scenario →
How fast is Grok 4.3?
How much does Grok 4.3 cost at scale?
| Tokens / month | Est. cost (blended 3:1) |
|---|---|
| 100,000 | $0.16 |
| 1,000,000 | $1.56 |
| 10,000,000 | $15.63 |
| 100,000,000 | $156.25 |
How does Grok 4.3 compare with other models?
What is Grok 4.3 best for?
What should you explore next for Grok 4.3?
Which Grok 4.3 head-to-head comparisons are available?
What are common questions about Grok 4.3?
Is Grok 4.3 cheaper than Gemini 3.7 Flash?
Grok 4.3 costs $1.56/M blended tokens, Gemini 3.7 Flash costs $1.50/M — Gemini 3.7 Flash is cheaper.
How much does 1 million tokens cost with Grok 4.3?
At a 3:1 input:output ratio, 1 million blended tokens costs approximately $1.56. Pure input costs $1.25/M; pure output costs $2.50/M.
What does Grok 4.3 cost at high volume?
At 100 million blended tokens a month, Grok 4.3 costs approximately $156.25. See the cost-at-scale table below for other volumes.