grok-4.20-0309-non-reasoning
xAIdocs.x.ai/developers/models/grok-4.20-0309-non-r
Grok 4.20 is a high-performance model with industry-leading speed and agentic tool calling capabilities. It combines the lowest hallucination rate on the market with strict prompt adherence, delivering consistently precise and truthful responses.
Updated 5 h ago · first seen 11 Sept 2026
model_01M299WBFFVB1CJG1KQZ98Q8E0
- Context
- 1M tokens
- T1 · 5 h ago
Specification
- Family
- Grok
Source:xAI — docsT1observed 5 h agohigh
- Status
- deprecated
Source:Artificial AnalysisT2observed 5 h agomedium
- Openness
- proprietary
Source:Artificial AnalysisT2observed 5 h agomedium
- Context window
- 1M tokens
Source:xAI — docsT1observed 5 h agohigh
- Modalities
- image, text
Source:xAI — docsT1observed 5 h agohigh
- Input modalities
- text, image
Source:xAI — docsT1observed 5 h agohigh
- Output modalities
- text
Source:xAI — docsT1observed 5 h agohigh
- API model id
- grok-4.20-0309-non-reasoning
Source:xAI — docsT1observed 5 h agohigh
- Official page
Source:xAI — docsT1observed 5 h agohigh
- Batch api
- Yes
Source:xAI — docsT1observed 5 h agohigh
- Aa median output tokens per second
- 94.3
Source:Artificial AnalysisT2observed 5 h agomedium
- Regions
- us-east-1, us-west-2
Source:xAI — docsT1observed 5 h agohigh
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
17
Source tiers
T1T214 / 3
Freshest observation
5 h ago
Conflicts
1 flagged
Modalities
- Modalities
- imagetext
- Input
- textimage
- Output
- text
Capabilities
Tool calling
Yes
xAI — docs · T1
Structured output
Yes
xAI — docs · T1
Reasoning
No
xAI — docs · T1
Vision
Yes
xAI — docs · T1
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
- 1M tokens
Source:xAI — docsT1observed 5 h agohigh
| Benchmark | Score | Metric | Config | Evaluated | Source |
|---|---|---|---|---|---|
| τ²-bench | 59.94% | pass^1 | aa_slug=grok-4-20-non-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Artificial Analysis Intelligence Index | 14.2 | index | aa_slug=grok-4-20-non-reasoning · version=4.3 · estimated=true | — | artificialanalysis.aiT2 |
| GPQA | 77.58% | accuracy | aa_slug=grok-4-20-non-reasoning · variant=GPQA Diamond · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Humanity's Last Exam | 27.94% | accuracy | aa_slug=grok-4-20-non-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| IFBench | 49.32% | accuracy | aa_slug=grok-4-20-non-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| MMMU-Pro | 64.91% | accuracy | aa_slug=grok-4-20-non-reasoning · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
| Terminal-Bench | 16.67% | accuracy | aa_slug=grok-4-20-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — | artificialanalysis.aiT2 |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). Results with different configs are not directly comparable — see methodology.
Current prices
| Provider | Input / 1M | Output / 1M | Cached in | Batch in / out | Context | Observed | Source |
|---|---|---|---|---|---|---|---|
| xAI APIgrok-4.20-0309-non-reasoning | $1.25 | $2.5 | $0.20 | $1 / $2 | 1M | 3 h ago | docs.x.aiT1 |
USD per 1M tokens as published by each provider (USD). Rows are append-only: every change is kept in the history below.
Available through 1
Price history
Input price · USD / 1M tokens
Output price · USD / 1M tokens
- xAI API
No hardware estimate available
No lineage recorded
Papers 0
No papers linked yet.
Repositories 0
No repositories linked yet.
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Openness
Opennessopenness1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| proprietary | → current | current | Artificial AnalysisT2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
grok-4.20-0309-non-reasoning scores 16.67% on Terminal-Bench
artificial_analysisgrok-4.20-0309-non-reasoning scores 59.94% on τ²-bench
artificial_analysisgrok-4.20-0309-non-reasoning scores 64.91% on MMMU-Pro
artificial_analysisgrok-4.20-0309-non-reasoning scores 49.32% on IFBench
artificial_analysisgrok-4.20-0309-non-reasoning scores 27.94% on Humanity's Last Exam
artificial_analysisgrok-4.20-0309-non-reasoning scores 77.58% on GPQA
artificial_analysisgrok-4.20-0309-non-reasoning scores 14.2 on Artificial Analysis Intelligence Index
artificial_analysisxAI API lists grok-4.20-0309-non-reasoning at $1.25 in / $2.5 out per 1M tokens
xai
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| xAI — docs | docs.x.ai/developers/models/grok-4.20-0309-non-reasoning.md | model_page | T1· Official | 3 h ago | 1 |
| xAI — docs | docs.x.ai/developers/models | model_docs | T1· Official | 3 h ago | 3 |
| Artificial Analysis | artificialanalysis.ai/leaderboards/models | leaderboard | T2· Quality secondary | 5 h ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.