Agentic
Directional only- GPT-5.1-Codex
- 51.4
- Estimated · #54/151
- GPT-5.4 mini
- 42.4
- Estimated · #113/151
- Basis
- BenchAlign lane · 2 vs 6 public rows
- Reading
- Directional only
GPT-5.1-Codex vs GPT-5.4 mini
Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GPT-5.4 mini has the higher public score estimate, 62.63 versus 51.62, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
1 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.
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
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K 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
GPT-5.1-Codex is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
GPT-5.1-Codex and GPT-5.4 mini are scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.
Confidence: limited
Long documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
2 GPT-5.1-Codex only
1 Shared
18 GPT-5.4 mini only
- Shared results
- 1
- GPT-5.1-Codex only
- 2
- GPT-5.4 mini only
- 18
- Like-for-like categories
- 0 / 8
4 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- GPT-5.1-Codex
- 47.4
- Estimated · #89/183
- GPT-5.4 mini
- 42.9
- Supported · #125/183
- Basis
- BenchAlign lane · 1 vs 4 public rows
- Reading
- Directional only
Knowledge
Directional only- GPT-5.1-Codex
- 53.3
- Estimated · #69/181
- GPT-5.4 mini
- 56.3
- Supported · #53/181
- Basis
- BenchAlign lane · 0 vs 5 public rows
- Reading
- Directional only
Instruction following
Directional only- GPT-5.1-Codex
- 85.3
- #42/120
- GPT-5.4 mini
- 89.6
- #23/120
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Directional only
Reasoning
Not comparable- GPT-5.1-Codex
- 70.6
- Unranked · 2 rankable rows
- GPT-5.4 mini
- 73.5
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.1-Codex
- Not ranked
- GPT-5.4 mini
- 44.5
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.1-Codex
- Not ranked
- GPT-5.4 mini
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.1-Codex
- 66.8
- Unranked · 1 rankable row
- GPT-5.4 mini
- 57.2
- #31/48
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
| Category | GPT-5.1-Codex | GPT-5.4 mini | Basis | Reading |
|---|---|---|---|---|
| Agentic | 51.4Estimated · #54/151 | 42.4Estimated · #113/151 | Directional onlyBenchAlign lane · 2 vs 6 public rows | Directional only |
| Coding | 47.4Estimated · #89/183 | 42.9Supported · #125/183 | Directional onlyBenchAlign lane · 1 vs 4 public rows | Directional only |
| Knowledge | 53.3Estimated · #69/181 | 56.3Supported · #53/181 | Directional onlyBenchAlign lane · 0 vs 5 public rows | Directional only |
| Instruction following | 85.3#42/120 | 89.6#23/120 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 70.6Unranked · 2 rankable rows | 73.5Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | 44.5Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 66.8Unranked · 1 rankable row | 57.2#31/48 | 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) 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.
A shared-evidence shape is not available.
BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.
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
- GPT-5.1-Codex
- $0.00625
- 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
- GPT-5.1-Codex
- $0.0925
- 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
- GPT-5.1-Codex
- $0.15
- Fits in one request
- GPT-5.4 mini
- $0.075
- Fits in one request
GPT-5.4 mini has the lower modeled cost
Costs use the listed standard API rates.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationGPT-5.1-CodexGPT-5.4 mini
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.1-Codex
GPT-5.4 mini
API model ID
GPT-5.1-Codex
gpt-5.1-codex
OpenAI GPT-5.1-Codex model documentationGPT-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.
GPT-5.1-Codex
$0.125 per 1M cached input tokens
OpenAI GPT-5.1-Codex model documentationGPT-5.4 mini
$0.075 per 1M cached input tokens
OpenAI pricingDocumented inputs
GPT-5.1-Codex
Not sourced
GPT-5.4 mini
text, image
OpenAI model catalogDocumented outputs
GPT-5.1-Codex
Not sourced
GPT-5.4 mini
Provider availability
GPT-5.1-Codex
Not sourced
GPT-5.4 mini
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
GPT-5.1-Codex
Reasoning
GPT-5.4 mini
Reasoning
Weight access
GPT-5.1-Codex
Proprietary
GPT-5.4 mini
Proprietary
License
GPT-5.1-Codex
Proprietary
GPT-5.4 mini
Proprietary
Release date
GPT-5.1-Codex
2025-10-15
GPT-5.4 mini
2026-03-17
If you already use one of these models
- Deployment change
- Both entries list OpenAI as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
- Quality signal
- GPT-5.4 mini has the higher public score estimate, 62.63 versus 51.62, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0925 vs $0.051. Cache-heavy agent loop: $0.15 vs $0.075.
- Context tradeoff
- Both models list 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 evidence21 rows
Agentic
Gert Labs
GPT-5.1-Codex49.68%SourceGPT-5.4 mini—Not directly comparable
JobBench
GPT-5.1-Codex26.2%SourceGPT-5.4 mini—Not directly comparable
Terminal-Bench 2.0
GPT-5.1-Codex—GPT-5.4 mini60%SourceNot directly comparable
OSWorld-Verified
GPT-5.1-Codex—GPT-5.4 mini72.1%SourceNot directly comparable
MCP Atlas
GPT-5.1-Codex—GPT-5.4 mini57.7%SourceNot directly comparable
Toolathlon
GPT-5.1-Codex—GPT-5.4 mini42.9%SourceNot directly comparable
τ²-bench results
GPT-5.1-Codex—GPT-5.4 mini93.4%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
GPT-5.1-Codex—GPT-5.4 mini54.7%SourceNot directly comparable
Coding
Vibe Code Bench
Shared sourceGPT-5.1-Codex13.12%GPT-5.4 mini47.97%GPT-5.4 mini leads this result
FrontierCode 1.1 Main
GPT-5.1-Codex—GPT-5.4 mini27.0%SourceNot directly comparable
LiveCodeBench (Vals)
GPT-5.1-Codex—GPT-5.4 mini81.5%SourceNot directly comparable
SWE-bench (Vals)
GPT-5.1-Codex—GPT-5.4 mini73.0%SourceNot directly comparable
Knowledge
GPQA
GPT-5.1-Codex—GPT-5.4 mini88%SourceNot directly comparable
HLE
GPT-5.1-Codex—GPT-5.4 mini41.5%SourceNot directly comparable
HLE w/o tools
GPT-5.1-Codex—GPT-5.4 mini28.2%SourceNot directly comparable
GPQA Diamond (Vals)
GPT-5.1-Codex—GPT-5.4 mini83.1%SourceNot directly comparable
MMLU-Pro (Vals)
GPT-5.1-Codex—GPT-5.4 mini84.6%SourceNot directly comparable
Math
Multimodal
Frequently asked questions
Which is better, GPT-5.1-Codex or GPT-5.4 mini?
GPT-5.4 mini has the higher public score estimate, 62.63 versus 51.62, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GPT-5.1-Codex or GPT-5.4 mini?
GPT-5.1-Codex scores higher for coding on the public lane, 47.4 to 42.9. GPT-5.1-Codex 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, GPT-5.1-Codex or GPT-5.4 mini?
GPT-5.1-Codex scores higher for agentic tasks on the public lane, 51.4 to 42.4. GPT-5.1-Codex and GPT-5.4 mini are scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.
Which costs less, GPT-5.1-Codex or GPT-5.4 mini?
For the stated presets, chat costs $0.00625 on GPT-5.1-Codex and $0.003 on GPT-5.4 mini; repository review costs $0.0925 and $0.051; the cache-heavy agent loop costs $0.15 and $0.075. Costs use the listed standard API rates.
Which has the larger context window, GPT-5.1-Codex or GPT-5.4 mini?
Both models list the same context window, 400K.
Related comparisons
Last updated September 4, 2026
Watch GPT-5.1-Codex 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.