Agentic
Like-for-like- Claude Opus 4.7 (Adaptive)
- 57.4
- Supported · #31/152
- Grok 4.20
- 26.7
- Supported · #145/152
- Basis
- BenchAlign lane · 7 vs 4 public rows
- Reading
- Claude Opus 4.7 (Adaptive) leads
Claude Opus 4.7 (Adaptive) vs Grok 4.20
Updated September 10, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Claude Opus 4.7 (Adaptive) has the higher public score estimate, 69.1 versus 67.13, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
8 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Share or export
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Agentic work
Tool use, computer use, and multi-step task completion
Claude Opus 4.7 (Adaptive)
Claude Opus 4.7 (Adaptive) leads on the public agentic lane, 57.4 to 26.7, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
Long documents
Prompts that approach the documented context limit
Grok 4.20
Grok 4.20 has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
Grok 4.20
Grok 4.20 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Show secondary and unsupported calls
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Grok 4.20
Grok 4.20 has the lower estimated token cost for this stated workload. Claude Opus 4.7 (Adaptive) has no published cached-input rate, so cached tokens use its listed input rate. Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Grok 4.20
Grok 4.20 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Claude Opus 4.7 (Adaptive) is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
13 Claude Opus 4.7 (Adaptive) only
8 Shared
15 Grok 4.20 only
- Shared results
- 8
- Claude Opus 4.7 (Adaptive) only
- 13
- Grok 4.20 only
- 15
- Like-for-like categories
- 1 / 8
3 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Coding
Directional only- Claude Opus 4.7 (Adaptive)
- 57.6
- Estimated · #31/151
- Grok 4.20
- 28.2
- Supported · #141/151
- Basis
- BenchAlign lane · 3 vs 6 public rows
- Reading
- Directional only
Knowledge
Directional only- Claude Opus 4.7 (Adaptive)
- 61.1
- Estimated · #33/183
- Grok 4.20
- 49.4
- Supported · #85/183
- Basis
- BenchAlign lane · 4 vs 6 public rows
- Reading
- Directional only
Multimodal
Directional only- Claude Opus 4.7 (Adaptive)
- 50.3
- #36/48
- Grok 4.20
- 34.6
- #43/48
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Claude Opus 4.7 (Adaptive)
- 49.7
- Unranked · 3 rankable rows
- Grok 4.20
- 34.2
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 2 vs 2 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Opus 4.7 (Adaptive)
- Not ranked
- Grok 4.20
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude Opus 4.7 (Adaptive)
- Not ranked
- Grok 4.20
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude Opus 4.7 (Adaptive)
- Not ranked
- Grok 4.20
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Claude Opus 4.7 (Adaptive) | Grok 4.20 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 57.4Supported · #31/152 | 26.7Supported · #145/152 | Like-for-likeBenchAlign lane · 7 vs 4 public rows | Claude Opus 4.7 (Adaptive) leads |
| Coding | 57.6Estimated · #31/151 | 28.2Supported · #141/151 | Directional onlyBenchAlign lane · 3 vs 6 public rows | Directional only |
| Knowledge | 61.1Estimated · #33/183 | 49.4Supported · #85/183 | Directional onlyBenchAlign lane · 4 vs 6 public rows | Directional only |
| Multimodal | 50.3#36/48 | 34.6#43/48 | Directional onlyProvisional lane · 2 vs 2 weighted rows | Directional only |
| Reasoning | 49.7Unranked · 3 rankable rows | 34.2Unranked · 2 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
CharXiv
Multimodal
Claude Opus 4.7 (Adaptive): 91%Grok 4.20: 60.9%Normalized gap 30.1Claude Opus 4.7 (Adaptive) source Grok 4.20 sourceARC-AGI-2
Reasoning
Claude Opus 4.7 (Adaptive): 75.8%Grok 4.20: 53.3%Normalized gap 22.5Claude Opus 4.7 (Adaptive) source Grok 4.20 sourceTerminal-Bench 2.0
Agentic
Claude Opus 4.7 (Adaptive): 69.4%Grok 4.20: 47.1%Normalized gap 22.3Claude Opus 4.7 (Adaptive) source Grok 4.20 sourceHLE w/o tools
Knowledge
Claude Opus 4.7 (Adaptive): 46.9%Grok 4.20: 31.6%Normalized gap 15.3Claude Opus 4.7 (Adaptive) source Grok 4.20 sourceSWE-bench Pro
Coding
Claude Opus 4.7 (Adaptive): 64.3%Grok 4.20: 51.8%Normalized gap 12.5Claude Opus 4.7 (Adaptive) source Grok 4.20 source
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Claude Opus 4.7 (Adaptive)
- $0.0175
- Fits in one request
- Grok 4.20
- $0.005
- Fits in one request
Grok 4.20 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude Opus 4.7 (Adaptive)
- $0.325
- Fits in one request
- Grok 4.20
- $0.118
- Fits in one request
Grok 4.20 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Claude Opus 4.7 (Adaptive)
- $1.35
- Fits in one request
- Cached input priced at the published list-input rate
- Grok 4.20
- $0.5
- Fits in one request
- Cached input priced at the published list-input rate
Grok 4.20 has the lower modeled cost
Claude Opus 4.7 (Adaptive) has no published cached-input rate, so cached tokens use its listed input rate. Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationClaude Opus 4.7 (Adaptive)Grok 4.20
Context window
Maximum documented context; output-token limits may be lower.
Claude Opus 4.7 (Adaptive)
1M
Grok 4.20
2M
API model ID
Claude Opus 4.7 (Adaptive)
Not sourced
Grok 4.20
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Opus 4.7 (Adaptive)
Not published
Grok 4.20
Not published
Documented inputs
Claude Opus 4.7 (Adaptive)
Not sourced
Grok 4.20
Not sourced
Documented outputs
Claude Opus 4.7 (Adaptive)
Not sourced
Grok 4.20
Not sourced
Provider availability
Claude Opus 4.7 (Adaptive)
Not sourced
Grok 4.20
Not sourced
Reasoning profile
Claude Opus 4.7 (Adaptive)
Reasoning
Grok 4.20
Reasoning
Weight access
Claude Opus 4.7 (Adaptive)
Proprietary
Grok 4.20
Proprietary
License
Claude Opus 4.7 (Adaptive)
Proprietary
Grok 4.20
Proprietary
Release date
Claude Opus 4.7 (Adaptive)
2026-04-16
Grok 4.20
2026-03-10
If you already use one of these models
- Deployment change
- The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
- Quality signal
- Claude Opus 4.7 (Adaptive) has the higher public score estimate, 69.1 versus 67.13, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.325 vs $0.118. Cache-heavy agent loop: $1.35 vs $0.5.
- Context tradeoff
- Grok 4.20 has the larger documented window (2M).
Run the same representative tasks against both endpoints before changing production traffic.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence36 rows
Agentic
Terminal-Bench 2.0
Claude Opus 4.7 (Adaptive)69.4%SourceGrok 4.2047.1%SourceClaude Opus 4.7 (Adaptive) leads this result
BrowseComp
Claude Opus 4.7 (Adaptive)79.3%SourceGrok 4.20—Not directly comparable
MCP Atlas
Claude Opus 4.7 (Adaptive)77.3%SourceGrok 4.20—Not directly comparable
OSWorld-Verified
Claude Opus 4.7 (Adaptive)78%SourceGrok 4.20—Not directly comparable
CyberGym
Claude Opus 4.7 (Adaptive)73.1%SourceGrok 4.20—Not directly comparable
OSWorld 2.0
Claude Opus 4.7 (Adaptive)18.2%SourceGrok 4.20—Not directly comparable
JobBench
Claude Opus 4.7 (Adaptive)45.9%SourceGrok 4.20—Not directly comparable
DeepSearchQA
Claude Opus 4.7 (Adaptive)—Grok 4.2062.8%SourceNot directly comparable
Gert Labs
Claude Opus 4.7 (Adaptive)—Grok 4.2038.36%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Claude Opus 4.7 (Adaptive)—Grok 4.2044.2%SourceNot directly comparable
Coding
SWE-bench Verified
Claude Opus 4.7 (Adaptive)87.6%SourceGrok 4.2076.7%SourceClaude Opus 4.7 (Adaptive) leads this result
SWE-bench Pro
Claude Opus 4.7 (Adaptive)64.3%SourceGrok 4.2051.8%SourceClaude Opus 4.7 (Adaptive) leads this result
Terminal-Bench 2.0
Claude Opus 4.7 (Adaptive)69.4%SourceGrok 4.20—Not directly comparable
LiveCodeBench Pro
Claude Opus 4.7 (Adaptive)—Grok 4.2074.2%SourceNot directly comparable
Vibe Code Bench
Claude Opus 4.7 (Adaptive)—Grok 4.204.06%SourceNot directly comparable
LiveCodeBench (Vals)
Claude Opus 4.7 (Adaptive)—Grok 4.2084.3%SourceNot directly comparable
SWE-bench (Vals)
Claude Opus 4.7 (Adaptive)—Grok 4.2072.2%SourceNot directly comparable
Reasoning
MRCR v2 128K-256K
Claude Opus 4.7 (Adaptive)59.2%SourceGrok 4.20—Not directly comparable
ARC-AGI-2
Claude Opus 4.7 (Adaptive)75.8%SourceGrok 4.2053.3%SourceClaude Opus 4.7 (Adaptive) leads this result
ARC-AGI-3
Shared sourceClaude Opus 4.7 (Adaptive)0.2%Grok 4.200.1%Claude Opus 4.7 (Adaptive) leads this result
Knowledge
GPQA
Claude Opus 4.7 (Adaptive)94.2%SourceGrok 4.20—Not directly comparable
GPQA-D
Claude Opus 4.7 (Adaptive)94.2%SourceGrok 4.2088.5%SourceClaude Opus 4.7 (Adaptive) leads this result
HLE
Claude Opus 4.7 (Adaptive)54.7%SourceGrok 4.20—Not directly comparable
HLE w/o tools
Claude Opus 4.7 (Adaptive)46.9%SourceGrok 4.2031.6%SourceClaude Opus 4.7 (Adaptive) leads this result
HealthBench Hard
Claude Opus 4.7 (Adaptive)—Grok 4.2020.3%SourceNot directly comparable
MedXpertQA (Text)
Claude Opus 4.7 (Adaptive)—Grok 4.2050.2%SourceNot directly comparable
GPQA Diamond (Vals)
Claude Opus 4.7 (Adaptive)—Grok 4.2088.6%SourceNot directly comparable
MMLU-Pro (Vals)
Claude Opus 4.7 (Adaptive)—Grok 4.2086.3%SourceNot directly comparable
Math
FrontierMath (legacy)
Claude Opus 4.7 (Adaptive)43.8%SourceGrok 4.20—Not directly comparable
Multimodal
OfficeQA Pro
Claude Opus 4.7 (Adaptive)43.6%SourceGrok 4.20—Not directly comparable
CharXiv
Claude Opus 4.7 (Adaptive)91%SourceGrok 4.2060.9%SourceClaude Opus 4.7 (Adaptive) leads this result
CharXiv w/o tools
Claude Opus 4.7 (Adaptive)82.1%SourceGrok 4.20—Not directly comparable
MMMU-Pro
Claude Opus 4.7 (Adaptive)—Grok 4.2075.2%SourceNot directly comparable
ERQA
Claude Opus 4.7 (Adaptive)—Grok 4.2054.1%SourceNot directly comparable
SimpleVQA
Claude Opus 4.7 (Adaptive)—Grok 4.2057.4%SourceNot directly comparable
MedXpertQA (MM)
Claude Opus 4.7 (Adaptive)—Grok 4.2065.8%SourceNot directly comparable
Frequently asked questions
Which is better, Claude Opus 4.7 (Adaptive) or Grok 4.20?
Claude Opus 4.7 (Adaptive) has the higher public score estimate, 69.1 versus 67.13, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Claude Opus 4.7 (Adaptive) or Grok 4.20?
Claude Opus 4.7 (Adaptive) scores higher for coding on the public lane, 57.6 to 28.2. Claude Opus 4.7 (Adaptive) is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which is better for agentic tasks, Claude Opus 4.7 (Adaptive) or Grok 4.20?
Claude Opus 4.7 (Adaptive) leads the public agentic tasks lane, 57.4 to 26.7, with Supported evidence for both models and non-overlapping 90% intervals.
Which costs less, Claude Opus 4.7 (Adaptive) or Grok 4.20?
For the stated presets, chat costs $0.0175 on Claude Opus 4.7 (Adaptive) and $0.005 on Grok 4.20; repository review costs $0.325 and $0.118; the cache-heavy agent loop costs $1.35 and $0.5. Claude Opus 4.7 (Adaptive) has no published cached-input rate, so cached tokens use its listed input rate. Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Claude Opus 4.7 (Adaptive) or Grok 4.20?
Grok 4.20 has the larger documented context window: 2M, compared with 1M.
Related comparisons
Last updated September 10, 2026
Watch Claude Opus 4.7 (Adaptive) vs Grok 4.20
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.