Updated October 2, 2026. Rank says Kimi K3 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
Kimi K3 has the higher public score estimate, 72.14 versus 68.91, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 16 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
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.
Code generation, repair, and software-engineering tasks
Kimi K3
Kimi K3 has the higher public coding point estimate, 61.4 to 49.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Tool use, computer use, and multi-step task completion
Kimi K3
Kimi K3 has the higher public agentic point estimate, 68.1 to 48.7, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
1K fresh input + 500 output tokens
GPT-5.4
GPT-5.4 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
GPT-5.4
GPT-5.4 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
50K fresh input + 3K output tokens
GPT-5.4
GPT-5.4 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
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.
Like-for-like · BenchAlign v5.8
Kimi K3 has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
1 category rests on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
ARC-AGI-2Reasoning
Normalized gap 13.6OfficeQA ProMultimodal
Normalized gap 10.1BrowseCompAgentic
Normalized gap 8.5CharXivMultimodal
Normalized gap 8.5HLEKnowledge
Normalized gap 3.9Each row shows the public-lane category score for both models: the BenchAlign v5.8 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.
| Category | GPT-5.4 | Kimi K3 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 48.7Supported · #43/119 | 68.1Supported · #8/119 | Like-for-likeBenchAlign v5.8 lane · 14 vs 12 public rows | Kimi K3 leads |
| Coding | 49.6Supported · #44/144 | 61.4Supported · #18/144 | Like-for-likeBenchAlign v5.8 lane · 5 vs 14 public rows | Kimi K3 leads · intervals overlap |
| Multimodal | 69.3#20/49 | 89.4#1/49 | Like-for-likeProvisional lane · 3 vs 3 weighted rows | Kimi K3 leads |
| Knowledge | 66.1Supported · #24/171 | 67.9Supported · #18/171 | Like-for-likeBenchAlign v5.8 lane · 7 vs 6 public rows | Kimi K3 leads · intervals overlap |
| Reasoning | 60.6#21/27 | 65.8#18/27 | Directional onlyProvisional lane · 2 vs 1 weighted rows | Directional only |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | 89.2#19/125 | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 64.4Unranked · 2 rankable rows | Not ranked | Not comparableProvisional lane · 2 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) 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.
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.
1K fresh input + 500 output tokens
GPT-5.4 has the lower modeled cost
Costs use the listed standard API rates.
50K fresh input + 3K output tokens
GPT-5.4 has the lower modeled cost
Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
GPT-5.4 has the lower modeled cost
Costs use the listed standard API rates.
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
Maximum documented context; output-token limits may be lower.
GPT-5.4
1.05M
OpenAI pricingKimi K3
1.05M
GPT-5.4
gpt-5.4
OpenAI pricingKimi K3
Not sourced
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.4
$0.25 per 1M cached input tokens
OpenAI pricingKimi K3
$0.3 per 1M cached input tokens
GPT-5.4
Not sourced
Kimi K3
Not sourced
GPT-5.4
Not sourced
Kimi K3
Not sourced
GPT-5.4
Not sourced
Kimi K3
Not sourced
GPT-5.4
Reasoning
Kimi K3
Reasoning
GPT-5.4
Proprietary
Kimi K3
Pending
GPT-5.4
Proprietary
Kimi K3
Pending
GPT-5.4
2026-03-05
Kimi K3
2026-07-16
Kimi K3 has the higher public score estimate, 72.14 versus 68.91, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Kimi K3 has the higher public coding point estimate, 61.4 to 49.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Kimi K3 has the higher public agentic tasks point estimate, 68.1 to 48.7, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
For the stated presets, chat costs $0.01 on GPT-5.4 and $0.0105 on Kimi K3; repository review costs $0.17 and $0.195; the cache-heavy agent loop costs $0.25 and $0.27. Costs use the listed standard API rates.
Both models list the same context window, 1.05M.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Terminal-Bench 2.0
Not directly comparable
CyberGym
Not directly comparable
BrowseComp
Kimi K3 leads this result
OSWorld-Verified
Not directly comparable
MCP Atlas
Kimi K3 leads this result
Toolathlon
Not directly comparable
τ²-bench results
Not directly comparable
Claw-Eval
Not directly comparable
DeepSearchQA
Kimi K3 leads this result
Gert Labs
Not directly comparable
ResearchClawBench
Not directly comparable
JobBench
Kimi K3 leads this result
ExploitGym
Not directly comparable
ApprenticeBench
Shared sourceKimi K3 leads this result
Terminal-Bench 2.1
Not directly comparable
Toolathlon-Verified
Not directly comparable
AutomationBench
Not directly comparable
APEX-Agents
Not directly comparable
SpreadsheetBench 2
Not directly comparable
DECK-Bench
Not directly comparable
Terminal-Bench 2.1 (Vals)
Not directly comparable
LiveCodeBench Pro
Not directly comparable
SWE-bench Pro
Not directly comparable
React Native Evals
Not directly comparable
Vibe Code Bench
Not directly comparable
PostTrainBench v1.1
Shared sourceKimi K3 leads this result
DeepSWE
Not directly comparable
CursorBench 3.2
Not directly comparable
FrontierSWE
Not directly comparable
ProgramBench
Not directly comparable
Kimi Code Bench v2
Not directly comparable
sweMarathon
Not directly comparable
PostTrain Bench
Not directly comparable
MLS-Bench Lite
Not directly comparable
VulcanBench v3
Not directly comparable
OpenHarmony Bench
Not directly comparable
FrontierSWE v2
Not directly comparable
LiveCodeBench (Vals)
Not directly comparable
SWE-bench (Vals)
Not directly comparable
MMMU-Pro
Kimi K3 leads this result
OfficeQA Pro
Kimi K3 leads this result
MMMU-Pro w/ Python
Kimi K3 leads this result
CharXiv
Kimi K3 leads this result
ERQA
Not directly comparable
SimpleVQA
Not directly comparable
ScreenSpot Pro
Not directly comparable
ZeroBench
GPT-5.4 leads this result
MedXpertQA (MM)
Not directly comparable
CharXiv w/o tools
Not directly comparable
MathVision
Not directly comparable
MathVision w/ Python
Not directly comparable
BabyVision w/ Python
Not directly comparable
ZeroBench w/ Python
Not directly comparable
WorldVQA ForceAnswer
Not directly comparable
OmniDocBench
Not directly comparable
PerceptionBench
Not directly comparable
GPQA
Kimi K3 leads this result
HLE
Kimi K3 leads this result
HLE w/o tools
Kimi K3 leads this result
GPQA-D
Kimi K3 leads this result
HealthBench Hard
Not directly comparable
MedXpertQA (Text)
Not directly comparable
HealthBench Professional
Not directly comparable
GPQA Diamond (Vals)
Not directly comparable
MMLU-Pro (Vals)
Not directly comparable
Gray Swan IPI (15 attempts)
Not directly comparable
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 October 2, 2026
