Coding work
Code generation, repair, and software-engineering tasks
Claude Sonnet 5
Claude Sonnet 5 leads on the public coding lane, 59.8 to 38.8, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
Claude Sonnet 5 vs MiniMax M3
Updated September 29, 2026. Rank says Claude Sonnet 5 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
Claude Sonnet 5 has the higher public score estimate, 66.99 versus 54.8, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 11 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 11
- Claude Sonnet 5 only
- 15
- MiniMax M3 only
- 16
- Like-for-like categories
- 3 / 8
Supported: Claude Sonnet 5 and MiniMax M3How 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
Claude Sonnet 5
Claude Sonnet 5 leads on the public agentic lane, 64.6 to 39.7, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: strongerChat turn cost
1K fresh input + 500 output tokens
MiniMax M3
MiniMax M3 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
MiniMax M3
MiniMax M3 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 M3
MiniMax M3 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-ratesLong documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
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.
59.8Claude Sonnet 538.8MiniMax M3
Like-for-like · BenchAlign v5.7
Claude Sonnet 5 leads the like-for-like coding row.
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.
15 Claude Sonnet 5 only
11 Shared
16 MiniMax M3 only
1 category rests on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
OSWorld-VerifiedAgentic
Normalized gap 11.1Claude Sonnet 5:81.2%MiniMax M3:70.1%SWE-bench VerifiedCoding
Normalized gap 4.7Claude Sonnet 5:85.2%MiniMax M3:80.5%SWE-bench ProCoding
Normalized gap 4.2Claude Sonnet 5:63.2%MiniMax M3:59%MMLU-Pro (Vals)Knowledge
Normalized gap 3.3Claude Sonnet 5:87.5%MiniMax M3:84.2%BrowseCompAgentic
Normalized gap 1.2Claude Sonnet 5:84.7%MiniMax M3:83.5%
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- Claude Sonnet 5
- 64.6
- Supported · #12/117
- MiniMax M3
- 39.7
- Supported · #56/117
- Basis
- BenchAlign v5.7 lane · 7 vs 9 public rows
- Reading
- Claude Sonnet 5 leads
Coding
Like-for-like- Claude Sonnet 5
- 59.8
- Supported · #19/142
- MiniMax M3
- 38.8
- Supported · #65/142
- Basis
- BenchAlign v5.7 lane · 11 vs 10 public rows
- Reading
- Claude Sonnet 5 leads
Knowledge
Like-for-like- Claude Sonnet 5
- 64.5
- Supported · #26/168
- MiniMax M3
- 48.3
- Supported · #67/168
- Basis
- BenchAlign v5.7 lane · 6 vs 2 public rows
- Reading
- Claude Sonnet 5 leads · intervals overlap
Multimodal
Directional only- Claude Sonnet 5
- 78.4
- #15/50
- MiniMax M3
- 52.0
- #37/50
- Basis
- Provisional lane · 1 vs 2 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- Claude Sonnet 5
- 78.7
- Unranked · 2 rankable rows
- MiniMax M3
- 79.4
- #5/27
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude Sonnet 5
- Not ranked
- MiniMax M3
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude Sonnet 5
- Not ranked
- MiniMax M3
- 92.4
- #5/124
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Claude Sonnet 5
- Not ranked
- MiniMax M3
- Not ranked
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
| Category | Claude Sonnet 5 | MiniMax M3 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 64.6Supported · #12/117 | 39.7Supported · #56/117 | Like-for-likeBenchAlign v5.7 lane · 7 vs 9 public rows | Claude Sonnet 5 leads |
| Coding | 59.8Supported · #19/142 | 38.8Supported · #65/142 | Like-for-likeBenchAlign v5.7 lane · 11 vs 10 public rows | Claude Sonnet 5 leads |
| Knowledge | 64.5Supported · #26/168 | 48.3Supported · #67/168 | Like-for-likeBenchAlign v5.7 lane · 6 vs 2 public rows | Claude Sonnet 5 leads · intervals overlap |
| Multimodal | 78.4#15/50 | 52.0#37/50 | Directional onlyProvisional lane · 1 vs 2 weighted rows | Directional only |
| Reasoning | 78.7Unranked · 2 rankable rows | 79.4#5/27 | 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 | 92.4#5/124 | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 1 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
- Claude Sonnet 5
- $0.007
- Fits in one request
- MiniMax M3
- $0.0009
- Fits in one request
MiniMax M3 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude Sonnet 5
- $0.13
- Fits in one request
- MiniMax M3
- $0.0186
- Fits in one request
MiniMax M3 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 Sonnet 5
- $0.18
- Fits in one request
- MiniMax M3
- $0.03
- Fits in one request
MiniMax M3 has the lower modeled cost
Costs use the listed standard API rates.
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.
SpecificationClaude Sonnet 5MiniMax M3
Context window
Maximum documented context; output-token limits may be lower.
Claude Sonnet 5
MiniMax M3
1M
API model ID
Claude Sonnet 5
claude-sonnet-5
Anthropic model overviewMiniMax M3
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude Sonnet 5
$0.2 per 1M cached input tokens
Claude API pricingMiniMax M3
$0.06 per 1M cached input tokens
Documented inputs
Claude Sonnet 5
text, image
Anthropic model overviewMiniMax M3
Not sourced
Documented outputs
Claude Sonnet 5
MiniMax M3
Not sourced
Provider availability
Claude Sonnet 5
Generally Available · Claude API
Anthropic model overviewMiniMax M3
Not sourced
Reasoning profile
Claude Sonnet 5
Reasoning
MiniMax M3
Non-Reasoning
Weight access
Claude Sonnet 5
Proprietary
MiniMax M3
Open Weight
License
Claude Sonnet 5
Proprietary
MiniMax M3
Open Weight
Release date
Claude Sonnet 5
2026-06-30
MiniMax M3
2026-06-01
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 Sonnet 5 has the higher public score estimate, 66.99 versus 54.8, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.13 vs $0.0186. Cache-heavy agent loop: $0.18 vs $0.03.
- Context tradeoff
- Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Claude Sonnet 5 or MiniMax M3?
Claude Sonnet 5 has the higher public score estimate, 66.99 versus 54.8, 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 Sonnet 5 or MiniMax M3?
Claude Sonnet 5 leads the public coding lane, 59.8 to 38.8, with Supported evidence for both models and non-overlapping 90% intervals.
Which is better for agentic tasks, Claude Sonnet 5 or MiniMax M3?
Claude Sonnet 5 leads the public agentic tasks lane, 64.6 to 39.7, with Supported evidence for both models and non-overlapping 90% intervals.
Which costs less, Claude Sonnet 5 or MiniMax M3?
For the stated presets, chat costs $0.007 on Claude Sonnet 5 and $0.0009 on MiniMax M3; repository review costs $0.13 and $0.0186; the cache-heavy agent loop costs $0.18 and $0.03. Costs use the listed standard API rates.
Which has the larger context window, Claude Sonnet 5 or MiniMax M3?
Both models list the same context window, 1M.
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence42 rows
Agentic
Terminal-Bench 3.0
Claude Sonnet 514.6%SourceMiniMax M3—Not directly comparable
Terminal-Bench 2.1
Claude Sonnet 580.4%SourceMiniMax M366.0%SourceClaude Sonnet 5 leads this result
BrowseComp
Claude Sonnet 584.7%SourceMiniMax M383.5%SourceClaude Sonnet 5 leads this result
HLE w/ tools
Claude Sonnet 557.4%SourceMiniMax M3—Not directly comparable
OSWorld-Verified
Claude Sonnet 581.2%SourceMiniMax M370.1%SourceClaude Sonnet 5 leads this result
Terminal-Bench 2.1 (Vals)
Claude Sonnet 574.5%SourceMiniMax M353.6%SourceClaude Sonnet 5 leads this result
ApprenticeBench
Claude Sonnet 516%SourceMiniMax M3—Not directly comparable
MCP Atlas
Claude Sonnet 5—MiniMax M374.2%SourceNot directly comparable
Claw-Eval
Claude Sonnet 5—MiniMax M374.5%SourceNot directly comparable
BankerToolBench
Claude Sonnet 5—MiniMax M376.1%SourceNot directly comparable
ResearchClawBench
Claude Sonnet 5—MiniMax M319.8%SourceNot directly comparable
OSWorld 2.0
Claude Sonnet 5—MiniMax M34.6%SourceNot directly comparable
Coding
SWE-bench Verified
Claude Sonnet 585.2%SourceMiniMax M380.5%SourceClaude Sonnet 5 leads this result
SWE-bench Pro
Claude Sonnet 563.2%SourceMiniMax M359%SourceClaude Sonnet 5 leads this result
SWE Multilingual
Claude Sonnet 578.3%SourceMiniMax M3—Not directly comparable
SWE Multimodal
Claude Sonnet 528.1%SourceMiniMax M3—Not directly comparable
Terminal-Bench 2.1
Claude Sonnet 580.4%SourceMiniMax M366.0%SourceClaude Sonnet 5 leads this result
FrontierCode 1.1 Main
Claude Sonnet 542.7%SourceMiniMax M3—Not directly comparable
cursorBench32
Claude Sonnet 561.5%SourceMiniMax M3—Not directly comparable
VulcanBench CII v1
Claude Sonnet 589.2%SourceMiniMax M3—Not directly comparable
LiveCodeBench (Vals)
Claude Sonnet 582.4%SourceMiniMax M382.2%SourceClaude Sonnet 5 leads this result
SWE-bench (Vals)
Claude Sonnet 579.6%SourceMiniMax M375.0%SourceClaude Sonnet 5 leads this result
cursorBench40
Claude Sonnet 534.1%SourceMiniMax M3—Not directly comparable
NL2Repo
Claude Sonnet 5—MiniMax M342.1%SourceNot directly comparable
VIBE V2
Claude Sonnet 5—MiniMax M350.1%SourceNot directly comparable
SVG-Bench
Claude Sonnet 5—MiniMax M363.7%SourceNot directly comparable
KernelBench Hard
Claude Sonnet 5—MiniMax M328.8%SourceNot directly comparable
OpenHarmony Bench
Claude Sonnet 5—MiniMax M348.4%SourceNot directly comparable
Multimodal
CharXiv
Claude Sonnet 588.3%SourceMiniMax M3—Not directly comparable
CharXiv w/o tools
Claude Sonnet 577%SourceMiniMax M3—Not directly comparable
OfficeQA Pro
Claude Sonnet 5—MiniMax M345.1%SourceNot directly comparable
OmniDocBench 1.5
Claude Sonnet 5—MiniMax M391.6%SourceNot directly comparable
MMMU-Pro
Claude Sonnet 5—MiniMax M378.1%SourceNot directly comparable
VideoMMMU
Claude Sonnet 5—MiniMax M384.6%SourceNot directly comparable
Video-MME (with subtitle)
Claude Sonnet 5—MiniMax M385.4%SourceNot directly comparable
Knowledge
HLE
Claude Sonnet 557.4%SourceMiniMax M3—Not directly comparable
HLE w/o tools
Claude Sonnet 543.2%SourceMiniMax M3—Not directly comparable
HLE-Verified
Claude Sonnet 531.0%SourceMiniMax M3—Not directly comparable
LABBench2
Claude Sonnet 580.1%SourceMiniMax M3—Not directly comparable
GPQA Diamond (Vals)
Claude Sonnet 588.9%SourceMiniMax M392.7%SourceMiniMax M3 leads this result
MMLU-Pro (Vals)
Claude Sonnet 587.5%SourceMiniMax M384.2%SourceClaude Sonnet 5 leads this result
Math
USAMO 2026
Claude Sonnet 5—MiniMax M385.7%SourceNot directly comparable
42 public results · 11 shared
Watch Claude Sonnet 5 vs MiniMax M3
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