Coding work
Code generation, repair, and software-engineering tasks
MiniMax M2.7
MiniMax M2.7 leads on the public coding lane, 36 to 22.5, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Grok 4.20 vs MiniMax M2.7
Updated September 29, 2026. Rank says Grok 4.20 is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty.
Both of these models will change. Get the price, version and retirement notices for the pair, each with its source. Follow model changes
Share or export
Decision reading
Grok 4.20 has the higher public score estimate, 59.74 versus 47.85, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 10 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 10
- Grok 4.20 only
- 13
- MiniMax M2.7 only
- 13
- Like-for-like categories
- 3 / 8
Estimated: Grok 4.20 · Supported: MiniMax M2.7How the comparison works
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
MiniMax M2.7
MiniMax M2.7 leads on the public agentic lane, 29.1 to 23.3, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limitedLong documents
Prompts that approach the documented context limit
Grok 4.20
Grok 4.20 has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Chat turn cost
1K fresh input + 500 output tokens
MiniMax M2.7
MiniMax M2.7 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesRepository review cost
50K fresh input + 3K output tokens
MiniMax M2.7
MiniMax M2.7 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
22.5Grok 4.2036.0MiniMax M2.7
Like-for-like · BenchAlign v5.7
MiniMax M2.7 leads the like-for-like coding row, although the 90% intervals overlap.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
13 Grok 4.20 only
10 Shared
13 MiniMax M2.7 only
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.
Terminal-Bench 2.0Agentic
Normalized gap 9.9Grok 4.20:47.1%MiniMax M2.7:57%MMLU-Pro (Vals)Knowledge
Normalized gap 5.9Grok 4.20:86.3%MiniMax M2.7:80.4%SWE-bench ProCoding
Normalized gap 4.4Grok 4.20:51.8%MiniMax M2.7:56.2%LiveCodeBench (Vals)Coding
Normalized gap 4.4Grok 4.20:84.3%MiniMax M2.7:79.9%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.7 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.
Agentic
Like-for-like- Grok 4.20
- 23.3
- Supported · #95/117
- MiniMax M2.7
- 29.1
- Supported · #82/117
- Basis
- BenchAlign v5.7 lane · 4 vs 7 public rows
- Reading
- MiniMax M2.7 leads · intervals overlap
Coding
Like-for-like- Grok 4.20
- 22.5
- Supported · #115/143
- MiniMax M2.7
- 36.0
- Supported · #77/143
- Basis
- BenchAlign v5.7 lane · 6 vs 11 public rows
- Reading
- MiniMax M2.7 leads · intervals overlap
Knowledge
Like-for-like- Grok 4.20
- 49.2
- Supported · #64/169
- MiniMax M2.7
- 43.0
- Supported · #86/169
- Basis
- BenchAlign v5.7 lane · 6 vs 4 public rows
- Reading
- Grok 4.20 leads · intervals overlap
Reasoning
Not comparable- Grok 4.20
- 41.1
- Unranked · 2 rankable rows
- MiniMax M2.7
- 76.1
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Grok 4.20
- 35.6
- #45/50
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 2 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Grok 4.20
- Not ranked
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Grok 4.20
- Not ranked
- MiniMax M2.7
- 91.6
- #10/124
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Grok 4.20
- Not ranked
- MiniMax M2.7
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Grok 4.20 | MiniMax M2.7 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 23.3Supported · #95/117 | 29.1Supported · #82/117 | Like-for-likeBenchAlign v5.7 lane · 4 vs 7 public rows | MiniMax M2.7 leads · intervals overlap |
| Coding | 22.5Supported · #115/143 | 36.0Supported · #77/143 | Like-for-likeBenchAlign v5.7 lane · 6 vs 11 public rows | MiniMax M2.7 leads · intervals overlap |
| Knowledge | 49.2Supported · #64/169 | 43.0Supported · #86/169 | Like-for-likeBenchAlign v5.7 lane · 6 vs 4 public rows | Grok 4.20 leads · intervals overlap |
| Reasoning | 41.1Unranked · 2 rankable rows | 76.1Unranked · 2 rankable rows | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
| Multimodal | 35.6#45/50 | Not ranked | Not comparableProvisional lane · 2 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 | 91.6#10/124 | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 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 v5.7) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
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
- Grok 4.20
- $0.0025
- Fits in one request
- MiniMax M2.7
- $0.0009
- Fits in one request
MiniMax M2.7 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Grok 4.20
- $0.07
- Fits in one request
- MiniMax M2.7
- $0.0186
- Fits in one request
MiniMax M2.7 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Grok 4.20
- $0.09
- Fits in one request
- MiniMax M2.7
- $0.078
- Does not fit in one request
- Cached input priced at the published list-input rate
MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGrok 4.20MiniMax M2.7
Context window
Maximum documented context; output-token limits may be lower.
Grok 4.20
1M
MiniMax M2.7
200K
API model ID
Grok 4.20
Not sourced
MiniMax M2.7
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Grok 4.20
$0.2 per 1M cached input tokens
MiniMax M2.7
Not published
Documented inputs
Grok 4.20
Not sourced
MiniMax M2.7
Not sourced
Documented outputs
Grok 4.20
Not sourced
MiniMax M2.7
Not sourced
Provider availability
Grok 4.20
Not sourced
MiniMax M2.7
Not sourced
Reasoning profile
Grok 4.20
Reasoning
MiniMax M2.7
Non-Reasoning
Weight access
Grok 4.20
Proprietary
MiniMax M2.7
Open Weight
License
Grok 4.20
Proprietary
MiniMax M2.7
Open Weight
Release date
Grok 4.20
2026-03-10
MiniMax M2.7
2026-03-18
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
- Grok 4.20 has the higher public score estimate, 59.74 versus 47.85, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.07 vs $0.0186. Cache-heavy agent loop: $0.09 vs $0.078.
- Context tradeoff
- Grok 4.20 has the larger documented window (1M).
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Grok 4.20 or MiniMax M2.7?
Grok 4.20 has the higher public score estimate, 59.74 versus 47.85, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, Grok 4.20 or MiniMax M2.7?
MiniMax M2.7 leads the public coding lane, 36 to 22.5, with Supported evidence for both models, although the 90% intervals overlap.
Which is better for agentic tasks, Grok 4.20 or MiniMax M2.7?
MiniMax M2.7 leads the public agentic tasks lane, 29.1 to 23.3, with Supported evidence for both models, although the 90% intervals overlap.
Which costs less, Grok 4.20 or MiniMax M2.7?
For the stated presets, chat costs $0.0025 on Grok 4.20 and $0.0009 on MiniMax M2.7; repository review costs $0.07 and $0.0186; the cache-heavy agent loop costs $0.09 and $0.078. MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Grok 4.20 or MiniMax M2.7?
Grok 4.20 has the larger documented context window: 1M, compared with 200K.
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
Grok 4.2047.1%SourceMiniMax M2.757%SourceMiniMax M2.7 leads this result
DeepSearchQA
Grok 4.2062.8%SourceMiniMax M2.7—Not directly comparable
Gert Labs
Shared sourceGrok 4.2038.36%MiniMax M2.740.40%MiniMax M2.7 leads this result
Terminal-Bench 2.1 (Vals)
Grok 4.2044.2%SourceMiniMax M2.748.7%SourceMiniMax M2.7 leads this result
Toolathlon
Grok 4.20—MiniMax M2.746.3%SourceNot directly comparable
MLE-Bench Lite
Grok 4.20—MiniMax M2.766.6%SourceNot directly comparable
MM-ClawBench
Grok 4.20—MiniMax M2.762.7%SourceNot directly comparable
Claw-Eval
Grok 4.20—MiniMax M2.748.7%SourceNot directly comparable
Coding
LiveCodeBench Pro
Grok 4.2074.2%SourceMiniMax M2.7—Not directly comparable
SWE-bench Verified
Grok 4.2076.7%SourceMiniMax M2.7—Not directly comparable
SWE-bench Pro
Grok 4.2051.8%SourceMiniMax M2.756.2%SourceMiniMax M2.7 leads this result
Vibe Code Bench
Shared sourceGrok 4.204.06%MiniMax M2.727.04%MiniMax M2.7 leads this result
LiveCodeBench (Vals)
Grok 4.2084.3%SourceMiniMax M2.779.9%SourceGrok 4.20 leads this result
SWE-bench (Vals)
Grok 4.2072.2%SourceMiniMax M2.773.8%SourceMiniMax M2.7 leads this result
SWE-bench Verified*
Grok 4.20—MiniMax M2.775.4%SourceNot directly comparable
SWE-Rebench
Grok 4.20—MiniMax M2.751.9%SourceNot directly comparable
SWE Multilingual
Grok 4.20—MiniMax M2.776.5%SourceNot directly comparable
Multi-SWE Bench
Grok 4.20—MiniMax M2.752.7%SourceNot directly comparable
VIBE-Pro
Grok 4.20—MiniMax M2.755.6%SourceNot directly comparable
NL2Repo
Grok 4.20—MiniMax M2.739.8%SourceNot directly comparable
React Native Evals
Grok 4.20—MiniMax M2.771.4%SourceNot directly comparable
Reasoning
Multimodal
MMMU-Pro
Grok 4.2075.2%SourceMiniMax M2.7—Not directly comparable
CharXiv
Grok 4.2060.9%SourceMiniMax M2.7—Not directly comparable
ERQA
Grok 4.2054.1%SourceMiniMax M2.7—Not directly comparable
SimpleVQA
Grok 4.2057.4%SourceMiniMax M2.7—Not directly comparable
MedXpertQA (MM)
Grok 4.2065.8%SourceMiniMax M2.7—Not directly comparable
Knowledge
GPQA-D
Grok 4.2088.5%SourceMiniMax M2.787.0%SourceGrok 4.20 leads this result
HLE w/o tools
Grok 4.2031.6%SourceMiniMax M2.7—Not directly comparable
HealthBench Hard
Grok 4.2020.3%SourceMiniMax M2.7—Not directly comparable
MedXpertQA (Text)
Grok 4.2050.2%SourceMiniMax M2.7—Not directly comparable
GPQA Diamond (Vals)
Grok 4.2088.6%SourceMiniMax M2.786.6%SourceGrok 4.20 leads this result
MMLU-Pro (Vals)
Grok 4.2086.3%SourceMiniMax M2.780.4%SourceGrok 4.20 leads this result
MMLU-Pro (Arcee)
Grok 4.20—MiniMax M2.780.8%SourceNot directly comparable
Math
AIME25 (Arcee)
Grok 4.20—MiniMax M2.780.0%SourceNot directly comparable
36 public results · 10 shared
Watch Grok 4.20 vs MiniMax M2.7
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.
Last updated September 29, 2026