Data as of October 27, 2026 · How the score is built
Released Mar 10, 20261M context
Grok 4.20
Decision readingGrok 4.20 scores 57.6 out of 100 and ranks #63 of 230. This profile shows 24 source-displayable benchmark rows; its strongest eligible category is Multimodal & Grounded at #50. API pricing is $1.25 input and $2.5 output per million tokens, with cached input at $0.2.
Released Mar 10, 2026 — see all recent releases
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
57.6/100
field median 50.2#63 of 230 ranked models
Public
#63of 230
Verified —
Price
$1.25input / $2.50 output
input median $0.75cached $0.20 · blended $1.88
Speed
108tok/s
field median 96 tok/sFirst token 17.04 s
Context
1Mtokens
field median 256,000Reported for this model; direct source link not stored
Strongest published evidence
Multimodal & Grounded ranks #50. Particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Validate before choosing
24 published rows leave some tracked benchmark slots empty. Coding is its lowest eligible category at #122.
Source-linked · 24 displayable benchmark rows
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #109 of 134Percentile 19thWeight 22%5 benchmarksMixed sources | 25.6 | 5 benchmarks | Mixed sources | |||
| CodingRank #122 of 155Percentile 21stWeight 20%6 benchmarksMixed sources | 21.8 | 6 benchmarks | Mixed sources | |||
| ReasoningRank Not rankedWeight 17%2 benchmarksMixed sources | 44.9 | 2 benchmarks | Mixed sources | |||
| MultimodalRank #50 of 58Percentile 14thWeight 12%5 benchmarksReported | 54.3 | 5 benchmarks | Reported | |||
| KnowledgeRank #72 of 189Percentile 62ndWeight 12%6 benchmarksMixed sources | 52.8 | 6 benchmarks | Mixed sources | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
24 of 774 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic3/5 verified
- Coding3/6 verified
- Reasoning1/2 verified
- Multimodal0/5 verified
- Knowledge2/6 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- MathNot measured
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
Grok 4.20 category percentile values
- Agentic19th percentile
- Coding21st percentile
- ReasoningNot eligible
- Multimodal14th percentile
- Knowledge62nd percentile
- MultilingualNot eligible
- Instruction followingNot eligible
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#109/134
- Coding#122/155
- ReasoningNot ranked
- Multimodal#50/58
- Knowledge#72/189
- MultilingualNot ranked
- Inst. FollowingNot ranked
- MathNot ranked
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SWE-bench Pro | Score51.8% | Versus best verified row Best verified: Claude Opus 5.5 · 89.9% | Gap38.1 behind | Weight26% ref. weight | Secondary exact |
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | Score84.3% | Versus best verified row Best verified: Claude Fable 5.1 · 90.5% | Gap6.2 behind | Weight8% ref. weight | |
| LiveCodeBench Pro | Score74.2% | Versus best verified row Best verified: Sakana Fugu-Ultra · 90.8% | Gap16.6 behind | WeightDisplay only | Secondary exact |
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score76.7% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap19.3 behind | WeightDisplay only | Secondary exact |
| Vibe Code BenchVibe Code Bench v1.1 | Score4.06% | Versus best verified row Best verified: Gemini 4 Argon · 91.90% | Gap87.8 behind | WeightDisplay only | Benchmark exact |
| SWE-bench (Vals)SWE-bench, Vals AI run | Score72.2% | Versus best verified row Best verified: Claude Opus 5 · 97.0% | Gap24.8 behind | WeightDisplay only | Verified |
Agentic5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 | Score47.1% | Versus best verified row Best verified: GPT-5.5 · 82% | Gap34.9 behind | Weight7% ref. weight | Secondary exact |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | Score44.2% | Versus best verified row Best verified: GPT-6 Astra · 87.3% | Gap43.1 behind | Weight4% ref. weight | |
| DeepSearchQA | Score62.8% | Versus best verified row Best verified: Atria Dawn Preview · 96.0% | Gap33.2 behind | Weight2% ref. weight | Secondary exact |
| Gert LabsGert Labs Composite Game Benchmark | Score38.36% | Versus best verified row Best verified: Claude Opus 4.8 · 72.97% | Gap34.6 behind | WeightDisplay only | Benchmark exact |
| MazeBenchMazeBench: visual spatial reasoning in a 3D open world | Score0% | Versus best verified row Best verified: GPT-6 Astra · 23% | Gap23 behind | WeightDisplay only | Benchmark exact |
Reasoning2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | Score53.3% | Versus best verified row Best verified: GPT-6 Astra · 95% | Gap41.7 behind | WeightWeighted 25% | Secondary exact |
| ARC-AGI-3Abstraction and Reasoning Corpus for AGI v3 | Score0.1% | Versus best verified row Best verified: GPT-6 Astra · 62.7% | Gap62.6 behind | WeightWeighted 15% | Benchmark exact |
Multimodal5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMMU-ProMassive Multi-discipline Multimodal Understanding Pro | Score75.2% | Versus best verified row Best verified: GPT-6.1 Terra · 85.5% | Gap10.3 behind | WeightWeighted 40% | Secondary exact |
| CharXivCharXiv Reasoning | Score60.9% | Versus best verified row Best verified: Qwen3.8 Max · 93.5% | Gap32.6 behind | WeightWeighted 20% | Secondary exact |
| ERQA | Score54.1% | Versus best verified row Best verified: Qwen3.8 Max · 77.8% | Gap23.7 behind | WeightDisplay only | Secondary exact |
| SimpleVQA | Score57.4% | Versus best verified row Best verified: Qwen3.7 Plus · 81.7% | Gap24.3 behind | WeightDisplay only | Secondary exact |
| MedXpertQA (MM)MedXpertQA Multimodal | Score65.8% | Versus best verified row Best verified: Qwen3.8 Max · 80.4% | Gap14.6 behind | WeightDisplay only | Secondary exact |
Knowledge6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HLE w/o toolsHumanity's Last Exam without tools | Score31.6% | Versus best verified row Best verified: Claude Opus 5.5 · 64.4% | Gap32.8 behind | Weight7% ref. weight | Secondary exact |
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | Score86.3% | Versus best verified row Best verified: Claude Fable 5.1 · 92.4% | Gap6.1 behind | Weight6% ref. weight | Verified |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | Score88.6% | Versus best verified row Best verified: Gemini 3.1 Pro · 95.5% | Gap6.9 behind | Weight2% ref. weight | |
| GPQA-DGPQA Diamond | Score88.5% | Versus best verified row Best verified: GPT-6 Astra · 96.0% | Gap7.5 behind | WeightDisplay only | Secondary exact |
| HealthBench Hard | Score20.3% | Versus best verified row Best verified: Muse Spark · 42.8% | Gap22.5 behind | WeightDisplay only | Secondary exact |
| MedXpertQA (Text)MedXpertQA Text | Score50.2% | Versus best verified row Best verified: Muse Spark · 52.6% | Gap2.4 behind | WeightDisplay only | Secondary exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 24 rowsWhat it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
Grok 4.20 · 57.6 score · $1.88 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Reasoning
Grok 4.20 Multi-agentSpec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not published
- Context window
- 1M
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not disclosed by the provider
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Established
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.20 per million cached input tokens
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
Grok 4.20 ranks #63 of 230 on the public leaderboard with a score of 57.6/100. It does not yet have enough sourced coverage for a verified position.
Grok 4.20 is a proprietary model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Tracked as unresolved. xAI's current docs now present this model as Grok 4.20, but a clean exact-value benchmark table is still missing. BenchLM also stores secondary exact values from Meta AI's April 8, 2026 comparison chart, which labels the same family as Grok 4.2 Reasoning.
Grok 4.20 sits in the Grok 4.20 family with Grok 4.20 Multi-agent. Its explicit predecessor is Grok 4.1. 24 of 774 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Multimodal & Grounded at #50, while its lowest eligible position is Coding at #122. particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Last updated October 27, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Where to run Grok 4.20 by API
OpenRouter is the one host we track that publishes a per-token price for it. Prices read from each host’s own pages on October 20, 2026.
| Host | Input /1M | Output /1M | Cached /1M | Context |
|---|---|---|---|---|
| OpenRouter | $1.25 | $2.50 | $0.20 | 2M |
Questions
How does Grok 4.20 perform overall in AI benchmarks?
Grok 4.20 ranks #63 out of 230 models on the public BenchAlign leaderboard, with a score of 57.6/100. Its evidence status is Estimated, and this profile shows 24 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Grok 4.20 good for knowledge and understanding?
Grok 4.20 ranks #72 out of 189 eligible models for knowledge and understanding, with a public category score of 52.8/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Grok 4.20 good for coding and programming?
Grok 4.20 ranks #122 out of 155 eligible models for coding and programming, with a public category score of 21.8/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Grok 4.20 good for reasoning and logic?
Grok 4.20 has source-displayable benchmark coverage for reasoning and logic, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Grok 4.20 good for agentic tool use and computer tasks?
Grok 4.20 ranks #109 out of 134 eligible models for agentic tool use and computer tasks, with a public category score of 25.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Grok 4.20 good for multimodal and grounded tasks?
Grok 4.20 ranks #50 out of 58 eligible models for multimodal and grounded tasks, with a public category score of 54.3/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Which sibling models are related to Grok 4.20?
Grok 4.20 belongs to the Grok 4.20 family. Related tracked variants include Grok 4.20 Multi-agent. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Grok 4.20 have full benchmark coverage on BenchLM?
No. Grok 4.20 currently has 25 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Grok 4.20?
Grok 4.20 has a reported context window of 1M in the exact-model catalog record. The value stays visible, but the profile marks its source link as unavailable instead of presenting it as directly documented. Maximum output length remains separate because providers often publish a different limit.