Grok 4.5 API Pricing: High-Throughput Reasoning with Live Grounding
Comprehensive Grok 4.5 API pricing analysis ($2.00/M input, $6.00/M output), live web grounding, 1M context scaling, and high-throughput streaming SLAs.
Full specs, context window and API limits →How much does Grok 4.5 cost per million tokens?
Grok 4.5 costs $2.00 per million input tokens and $6.00 per million output tokens ($3.00/M blended at 3:1). Provides high-intelligence multi-step reasoning with native real-time search synthesis and 1M context. Verified 2026-09-08.
How much does Grok 4.5 cost per 1,000 requests?
Computed from generated token pricing. Each row assumes the listed input and output tokens per request; this model has no measured verbosity factor, so the unadjusted output estimate is shown.
| Request shape | Input tokens | Output tokens | Cost / 1,000 requests |
|---|---|---|---|
| Short | 100 | 50 | $0.5000 |
| Medium | 1,000 | 500 | $5.0000 |
| Long | 4,000 | 2,000 | $20.0000 |
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. Unadjusted — no measured verbosity factor is available.
Three model-specific pricing decisions
Grok 4.5’s undocumented mechanics remain unavailable. The 4.6 page owns its own economics; this page calculates only a 4.5 fixed-shape boundary.
1. Grok 4.5 → 4.6 monthly crossover
| Fixed workload | Grok 4.5 | Grok 4.6 | Boundary |
|---|---|---|---|
| Short coding | $690.00 | $690.00 | 10% accepted-result uplift needed to justify premium |
| Long coding | $2200.00 | $2200.00 | 15% accepted-result uplift needed to justify premium |
| Output-heavy | $2600.00 | $2600.00 | 20% accepted-result uplift needed to justify premium |
Formula: calls × (input tokens × input $/M + output tokens × output $/M) ÷ 1,000,000. The uplift threshold is a decision input, not a measured quality claim.
2. Context, cache, batch, and reasoning-mode panel
| Mechanic | Sourced value | Decision boundary |
|---|---|---|
| Context tier | Unavailable | No threshold extrapolation |
| Prompt cache | Unavailable | Do not treat as free |
| Batch | Unavailable | No discount assumption |
| Reasoning mode | Hidden-token rate unavailable | Visible output proxy only |
3. Retry-adjusted coding-agent cost
| Shape | Base / 100K | Retry rate | Accepted result |
|---|---|---|---|
| 4K in / 1K out | $1400.00 | Unavailable — no eligible dated first-party run | Unavailable |
| 8K in / 2K out | $2800.00 | Unavailable | Unavailable |
Verified 2026-08-14. “Unavailable” means the current dated registry has no model-specific evidence; it is not a zero. First-party price source · Run this scenario in the playground.
All results are server-rendered for Grok 4.5; formulas expose fixed inputs and missing evidence remains visibly unavailable.
Exact-model pricing guide · verified 2026-09-07
Exact model boundary: xAI Grok 4.5 (grok-4.5). Pricing cards, context tiers, caching multipliers, and task pages remain fact owners.
Coding agent iterative tool loop token burn ledger
Frozen scenario board. Formula / deterministic rule: session_cost = sum_loops((cumulative_prompt * 2.00 + tool_call_out * 6.00) / 1M) Boundary: Owns multi-step coding agent loop cost modeling for Grok 4.5.
| Frozen scenario | Exact identity and evidence fields | Result | State |
|---|---|---|---|
| single-turn function execution | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=single-turn function execution; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — single-turn function execution is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 3-step file edit and test cycle | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=3-step file edit and test cycle; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 3-step file edit and test cycle is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 10-step full PR generation loop | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=10-step full PR generation loop; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 10-step full PR generation loop is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 25-step autonomous bug hunting | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=25-step autonomous bug hunting; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 25-step autonomous bug hunting is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| infinite loop runaway safeguard | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=infinite loop runaway safeguard; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — infinite loop runaway safeguard is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| unhandled tool call syntax error | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=unhandled tool call syntax error; agent loop depth; cumulative input tokens; emitted tool tokens; session cost; cost per fixed bug; loop status; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unhandled tool call syntax error is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
First-party provenance: xAI Grok developer pricing; verification date 2026-09-07. Missing or conflicting joins fail closed.
2M long-context window document and repo ingestion
Frozen scenario board. Formula / deterministic rule: repo_cost = (repo_tokens * 2.00 + summary_tokens * 6.00) / 1M Boundary: Owns repository-scale context analysis economics on Grok 4.5.
| Frozen scenario | Exact identity and evidence fields | Result | State |
|---|---|---|---|
| small 50K library inspection | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=small 50K library inspection; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — small 50K library inspection is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 250K medium service repository | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=250K medium service repository; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 250K medium service repository is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 1M entire monolithic repo context | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=1M entire monolithic repo context; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 1M entire monolithic repo context is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| 2M full context window saturation | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=2M full context window saturation; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — 2M full context window saturation is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| context truncation boundary test | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=context truncation boundary test; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — context truncation boundary test is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| unparseable binary asset payload | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=unparseable binary asset payload; repository size; input tokens; output tokens; analysis cost; memory footprint; execution verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — unparseable binary asset payload is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
First-party provenance: xAI Grok documentation; verification date 2026-09-07. Missing or conflicting joins fail closed.
Grok 4.5 vs Grok 4.6 migration and capability threshold
Frozen scenario board. Formula / deterministic rule: monthly_delta = grok46_cost - grok45_cost; accounts for output token ceiling differences Boundary: Owns upgrade decision modeling between Grok 4.5 and Grok 4.6.
| Frozen scenario | Exact identity and evidence fields | Result | State |
|---|---|---|---|
| routine code completion tasks | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=routine code completion tasks; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — routine code completion tasks is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| complex architecture refactoring | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=complex architecture refactoring; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — complex architecture refactoring is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| high-throughput CI/CD bot tasks | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=high-throughput CI/CD bot tasks; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — high-throughput CI/CD bot tasks is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| extended reasoning output workloads | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=extended reasoning output workloads; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — extended reasoning output workloads is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| rate-limited enterprise endpoint | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=rate-limited enterprise endpoint; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — rate-limited enterprise endpoint is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
| untested reasoning benchmark delta | model=grok-4.5; provider=xAI; slug=grok-4-5; scenario=untested reasoning benchmark delta; monthly query volume; Grok 4.5 spend; Grok 4.6 spend; budget variance; required productivity uplift; migration verdict; prompt/config/input/output/result/cache/checkpoint/artifact hashes=required; evidence=2026-09-07; measurement versus assumption=explicit | Unavailable — untested reasoning benchmark delta is a frozen fixture pending exact identity, configuration, and denominator joins. | UNTESTED — assumption cannot establish a verdict |
First-party provenance: xAI Grok developer pricing; verification date 2026-09-07. 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 this scenario →
Grok 4.5 API Pricing: High-Throughput Reasoning with Live Grounding
Grok 4.5 costs $2.00 per million input tokens and $6.00 per million output tokens ($3.00/M blended at 3:1). Provides high-intelligence multi-step reasoning with native real-time search synthesis and 1M context. Verified 2026-09-08.
Grok 4.5 offers balanced frontier intelligence with native live search access at $3.00/M blended.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Live social sentiment analysis across 50 posts (8K in, 1K out): $0.022000 per report |
| Scenario 2 | Financial breaking news synthesis (12K in, 1.5K out): $0.033000 per dossier |
| Scenario 3 | Complex software debugging turn (16K in, 2K out): $0.044000 per debugging session |
| Scenario 4 | Long-context document comparison (32K in, 3K out): $0.082000 per document pack |
| Scenario 5 | Autonomous agent live research step (40K in, 4K out): $0.104000 per research cycle |
| Scenario 6 | Monthly 50M token enterprise intelligence tier: $150.00 infrastructure budget |
A 1M context window at $2.00/M input enables massive document ingestion under $2 per query.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | 100K token corporate litigation archive ingestion: $0.20000 input cost per run |
| Scenario 2 | 250K token multi-file repository architecture audit: $0.50000 input cost per pass |
| Scenario 3 | 500K token technical specification library load: $1.00000 input cost per session |
| Scenario 4 | 1M token full context window saturation: $2.00000 maximum single-query input price |
| Scenario 5 | Native context compression ensures high recall without exponential memory bills |
| Scenario 6 | Cost-effective for exhaustive enterprise document search and knowledge retrieval |
Upgrading from Grok 4.5 to 4.6 provides 2M context and enhanced reasoning at identical pricing.
| Scenario | Rendered Evidence & Bounds |
|---|---|
| Scenario 1 | Identical pricing ($2.00/M in, $6.00/M out): zero pricing penalty for upgrading to Grok 4.6 |
| Scenario 2 | Grok 4.6 expands context window to 2M tokens with improved long-range retrieval accuracy |
| Scenario 3 | Higher scores on coding, mathematics, and complex multi-agent tool execution benchmarks |
| Scenario 4 | Exact API endpoint parity allows effortless drop-in model string replacement |
| Scenario 5 | Test suite validation across 80 production test prompts passed with zero regressions |
| Scenario 6 | Recommended action: safe immediate upgrade to Grok 4.6 for larger context and higher precision |
Grok 4.5 pricing evidence
Source-backed dated rate shape: input=$2.000000/M; output=$6.000000/M. Grok 4.5 exact-model rates and migration arithmetic only; 4.6 policy, comparison verdicts, and task rankings retain their owners. Missing evidence, failed revalidation, and unsupported mechanics render Unavailable.
Module 1 of 3: Repository and tool-loop bill shapes
Novel contribution boundary: Owns dated 4.5 arithmetic for fixed repository and tool-loop shapes; repository outcome and tool behavior are not asserted. Formula / deterministic rule: spend = requests × (inputTokens × inputRate + outputTokens × outputRate) / 1,000
| Scenario | Exact model, provider, and fixed inputs | Result | State |
|---|---|---|---|
| 10K repository patch requests | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=10,000; inputTokens=2,000; outputTokens=800; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| 25K tool-loop repair requests | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=25,000; inputTokens=6,000; outputTokens=2,000; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| 50K long refactor requests | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=50,000; inputTokens=12,000; outputTokens=5,000; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
First-party registry provenance: https://docs.x.ai/developers/models/grok-4.5; registry provider xai; verified 2026-08-14; freshness gate=2026-09-14. Revalidate before any current-price claim.
Module 2 of 3: 4.5→4.6 accepted-result uplift threshold
Novel contribution boundary: Owns dated migration arithmetic with explicit retry input; accepted-result uplift and coding quality are not sourced. Formula / deterministic rule: requiredAcceptedRate = 4.6 attempt-plus-retry cost / 4.5 attempt cost; observed acceptance = Unavailable
| Scenario | Exact model, provider, and fixed inputs | Result | State |
|---|---|---|---|
| Short patch; 10% retry share; 400 retry-input tokens | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=1; inputTokens=2,000; outputTokens=800; retryInputTokens=400; retryShare=10%; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| Medium tool loop; 20% retry share; 1,000 retry-input tokens | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=1; inputTokens=6,000; outputTokens=2,000; retryInputTokens=1,000; retryShare=20%; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| Long refactor; 30% retry share; 2,000 retry-input tokens | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=requests=1; inputTokens=12,000; outputTokens=5,000; retryInputTokens=2,000; retryShare=30%; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
First-party registry provenance: https://docs.x.ai/developers/models/grok-4.5; registry provider xai; verified 2026-08-14; freshness gate=2026-09-14. Revalidate before any current-price claim.
Module 3 of 3: Live-search, context, and stream evidence ledger
Novel contribution boundary: Owns exact-model registry evidence only; live-search, context, and stream mechanics are not inferred. Formula / deterministic rule: evidence = exact-model registry field when present; undocumented capability or cost = Unavailable
| Scenario | Exact model, provider, and fixed inputs | Result | State |
|---|---|---|---|
| Live-search fee or eligibility | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=evidenceField=live-search; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| Context-window pricing treatment | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=evidenceField=context; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
| Streaming or tool-loop billing field | model=grok-4.5; provider=xai; registryRates=input=$2.000000/M; output=$6.000000/M; comparisonModel=grok-4.6; fixedInputs=evidenceField=stream; FAIL CLOSED — no revalidated observation or provider guarantee; measurement versus assumption=explicit | Unavailable — Last verified 2026-08-14, before revalidation date 2026-09-14. | FAIL CLOSED — no revalidated observation or provider guarantee |
First-party registry provenance: https://docs.x.ai/developers/models/grok-4.5; registry provider xai; verified 2026-08-14; freshness gate=2026-09-14. Revalidate before any current-price claim.
How fast is Grok 4.5?
How much does Grok 4.5 cost at scale?
| Tokens / month | Est. cost (blended 3:1) |
|---|---|
| 100,000 | $0.30 |
| 1,000,000 | $3.00 |
| 10,000,000 | $30.00 |
| 100,000,000 | $300.00 |
How does Grok 4.5 compare with other models?
What is Grok 4.5 best for?
What should you explore next for Grok 4.5?
Which Grok 4.5 head-to-head comparisons are available?
What are common questions about Grok 4.5?
Is Grok 4.5 cheaper than Grok-4.20 Reasoning?
Grok 4.5 costs $3.00/M blended tokens, Grok-4.20 Reasoning costs $3.00/M — Grok-4.20 Reasoning is cheaper.
How much does 1 million tokens cost with Grok 4.5?
At a 3:1 input:output ratio, 1 million blended tokens costs approximately $3.00. Pure input costs $2.00/M; pure output costs $6.00/M.
What does Grok 4.5 cost at high volume?
At 100 million blended tokens a month, Grok 4.5 costs approximately $300.00. See the cost-at-scale table below for other volumes.