Math
Like-for-like- Claude 3.5 Sonnet
- 1.6
- GPT-5.4 mini
- 21.7
- Weighted basis
- 2 vs 2 rows
- Reading
- GPT-5.4 mini leads
Model comparison
Claude 3.5 Sonnet vs GPT-5.4 mini
Updated July 30, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
GPT-5.4 mini has the higher public score estimate, 55.79 versus 46.9, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
3 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.
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.
Long documents
Prompts that approach the documented context limit
GPT-5.4 mini
GPT-5.4 mini has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
GPT-5.4 mini
GPT-5.4 mini 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
Repository review cost
50K fresh input + 3K output tokens
GPT-5.4 mini
GPT-5.4 mini 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
No shared weighted benchmark basis supports a winner.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Cache-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. Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
1 Claude 3.5 Sonnet only
3 Shared
11 GPT-5.4 mini only
- Shared results
- 3
- Claude 3.5 Sonnet only
- 1
- GPT-5.4 mini only
- 11
- Like-for-like categories
- 1 / 8
1 category uses different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Category results, on a stated basis
Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.
Knowledge
Directional only- Claude 3.5 Sonnet
- 59.4
- GPT-5.4 mini
- 47.8
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Agentic
Not comparable- Claude 3.5 Sonnet
- Not measured
- GPT-5.4 mini
- 65.7
- Weighted basis
- 0 vs 2 rows
- Reading
- Not comparable
Coding
Not comparable- Claude 3.5 Sonnet
- 49.0
- GPT-5.4 mini
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Reasoning
Not comparable- Claude 3.5 Sonnet
- Not measured
- GPT-5.4 mini
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- Claude 3.5 Sonnet
- Not measured
- GPT-5.4 mini
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- Claude 3.5 Sonnet
- Not measured
- GPT-5.4 mini
- 76.6
- Weighted basis
- 0 vs 1 rows
- Reading
- Not comparable
Instruction following
Not comparable- Claude 3.5 Sonnet
- Not measured
- GPT-5.4 mini
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | Claude 3.5 Sonnet | GPT-5.4 mini | Weighted basis | Reading |
|---|---|---|---|---|
| Math | 1.6 | 21.7 | Like-for-like2 vs 2 rows | GPT-5.4 mini leads |
| Knowledge | 59.4 | 47.8 | Directional only1 vs 2 rows | Directional only |
| Agentic | Not measured | 65.7 | Not comparable0 vs 2 rows | Not comparable |
| Coding | 49.0 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Reasoning | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multimodal | Not measured | 76.6 | Not comparable0 vs 1 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
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.
GPQA
Knowledge
Claude 3.5 Sonnet: 59.4%GPT-5.4 mini: 88%Normalized gap 28.6Claude 3.5 Sonnet source GPT-5.4 mini sourceFrontierMath v2 (Tiers 1-3)
Math
FrontierMath v2 (Tier 4)
Math
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 3.5 Sonnet
- $0.0105
- Fits in one request
- GPT-5.4 mini
- $0.003
- Fits in one request
GPT-5.4 mini has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- Claude 3.5 Sonnet
- $0.195
- Fits in one request
- GPT-5.4 mini
- $0.051
- Fits in one request
GPT-5.4 mini 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 3.5 Sonnet
- $0.81
- Does not fit in one request
- Cached input priced at the published list-input rate
- GPT-5.4 mini
- $0.075
- Fits in one request
Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet 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 3.5 SonnetGPT-5.4 mini
Context window
Maximum documented context; output-token limits may be lower.
Claude 3.5 Sonnet
200K
GPT-5.4 mini
API model ID
Claude 3.5 Sonnet
Not sourced
GPT-5.4 mini
gpt-5.4-mini
OpenAI GPT-5.4 mini model documentationCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Claude 3.5 Sonnet
Not published
GPT-5.4 mini
$0.075 per 1M cached input tokens
OpenAI pricingDocumented inputs
Claude 3.5 Sonnet
Not sourced
GPT-5.4 mini
text, image
OpenAI model catalogDocumented outputs
Claude 3.5 Sonnet
Not sourced
GPT-5.4 mini
Provider availability
Claude 3.5 Sonnet
Not sourced
GPT-5.4 mini
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
Claude 3.5 Sonnet
Non-Reasoning
GPT-5.4 mini
Reasoning
Weight access
Claude 3.5 Sonnet
Proprietary
GPT-5.4 mini
Proprietary
License
Claude 3.5 Sonnet
Proprietary
GPT-5.4 mini
Proprietary
Release date
Claude 3.5 Sonnet
2024-06-01
GPT-5.4 mini
2026-03-17
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
- GPT-5.4 mini has the higher public score estimate, 55.79 versus 46.9, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.195 vs $0.051. Cache-heavy agent loop: $0.81 vs $0.075.
- Context tradeoff
- GPT-5.4 mini has the larger documented window (400K).
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 evidence15 rows
Agentic
Terminal-Bench 2.0
Claude 3.5 Sonnet—GPT-5.4 mini60%SourceNot directly comparable
OSWorld-Verified
Claude 3.5 Sonnet—GPT-5.4 mini72.1%SourceNot directly comparable
MCP Atlas
Claude 3.5 Sonnet—GPT-5.4 mini57.7%SourceNot directly comparable
Toolathlon
Claude 3.5 Sonnet—GPT-5.4 mini42.9%SourceNot directly comparable
τ²-bench results
Claude 3.5 Sonnet—GPT-5.4 mini93.4%SourceNot directly comparable
Coding
Knowledge
Math
FrontierMath v2 (Tiers 1-3)
Shared sourceClaude 3.5 Sonnet2.069%GPT-5.4 mini28.280%GPT-5.4 mini leads this result
FrontierMath v2 (Tier 4)
Shared sourceClaude 3.5 Sonnet0.000%GPT-5.4 mini2.080%GPT-5.4 mini leads this result
Multimodal
Frequently asked questions
Which is better, Claude 3.5 Sonnet or GPT-5.4 mini?
GPT-5.4 mini has the higher public score estimate, 55.79 versus 46.9, 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 3.5 Sonnet or GPT-5.4 mini?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, Claude 3.5 Sonnet or GPT-5.4 mini?
The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.
Which costs less, Claude 3.5 Sonnet or GPT-5.4 mini?
For the stated presets, chat costs $0.0105 on Claude 3.5 Sonnet and $0.003 on GPT-5.4 mini; repository review costs $0.195 and $0.051; the cache-heavy agent loop costs $0.81 and $0.075. Claude 3.5 Sonnet does not fit this workload in one request. Claude 3.5 Sonnet has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, Claude 3.5 Sonnet or GPT-5.4 mini?
GPT-5.4 mini has the larger documented context window: 400K, compared with 200K.
Related comparisons
Last updated July 30, 2026
Watch Claude 3.5 Sonnet vs GPT-5.4 mini
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.