Agentic
Like-for-like- GPT-5.6 Terra
- 87.4
- Kimi K3
- 89.5
- Weighted basis
- 2 vs 2 rows
- Reading
- Kimi K3 leads
GPT-5.6 Terra vs Kimi K3
Updated August 19, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
Kimi K3 has the higher public score estimate, 80.53 versus 72.94, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
9 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.
Agentic work
Tool use, computer use, and multi-step task completion
Kimi K3
Kimi K3 leads on the same 2 weighted benchmark rows.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
GPT-5.6 Terra
GPT-5.6 Terra 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-5.6 Terra
GPT-5.6 Terra 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-5.6 Terra
GPT-5.6 Terra 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
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.
15 GPT-5.6 Terra only
9 Shared
31 Kimi K3 only
- Shared results
- 9
- GPT-5.6 Terra only
- 15
- Kimi K3 only
- 31
- Like-for-like categories
- 1 / 8
2 categories use 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- GPT-5.6 Terra
- 92.9
- Kimi K3
- 61.0
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Multimodal
Directional only- GPT-5.6 Terra
- 80.7
- Kimi K3
- 78.5
- Weighted basis
- 1 vs 3 rows
- Reading
- Directional only
Coding
Not comparable- GPT-5.6 Terra
- 63.4
- Kimi K3
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Reasoning
Not comparable- GPT-5.6 Terra
- 83.9
- Kimi K3
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.6 Terra
- 80.8
- Kimi K3
- Not measured
- Weighted basis
- 2 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- GPT-5.6 Terra
- Not measured
- Kimi K3
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.6 Terra
- Not measured
- Kimi K3
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | GPT-5.6 Terra | Kimi K3 | Weighted basis | Reading |
|---|---|---|---|---|
| Agentic | 87.4 | 89.5 | Like-for-like2 vs 2 rows | Kimi K3 leads |
| Knowledge | 92.9 | 61.0 | Directional only1 vs 2 rows | Directional only |
| Multimodal | 80.7 | 78.5 | Directional only1 vs 3 rows | Directional only |
| Coding | 63.4 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Reasoning | 83.9 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Math | 80.8 | Not measured | Not comparable2 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 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.
BrowseComp
Agentic
Terminal-Bench 2.0
Agentic
MMMU-Pro
Multimodal
GPQA
Knowledge
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-5.6 Terra
- $0.008
- Fits in one request
- Kimi K3
- $0.0105
- Fits in one request
GPT-5.6 Terra has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- GPT-5.6 Terra
- $0.136
- Fits in one request
- Kimi K3
- $0.195
- Fits in one request
GPT-5.6 Terra 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-5.6 Terra
- $0.2
- Fits in one request
- Kimi K3
- $0.27
- Fits in one request
GPT-5.6 Terra 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-5.6 TerraKimi K3
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.6 Terra
1.05M
OpenAI model catalogKimi K3
1.05M
API model ID
GPT-5.6 Terra
gpt-5.6-terra
OpenAI model catalogKimi K3
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.6 Terra
$0.2 per 1M cached input tokens
OpenAI pricingKimi K3
$0.3 per 1M cached input tokens
Documented inputs
GPT-5.6 Terra
text, image
OpenAI model catalogKimi K3
Not sourced
Documented outputs
GPT-5.6 Terra
Kimi K3
Not sourced
Provider availability
GPT-5.6 Terra
Generally Available · OpenAI Responses API
OpenAI model catalogKimi K3
Not sourced
Reasoning profile
GPT-5.6 Terra
Reasoning
Kimi K3
Reasoning
Weight access
GPT-5.6 Terra
Proprietary
Kimi K3
Pending
License
GPT-5.6 Terra
Proprietary
Kimi K3
Pending
Release date
GPT-5.6 Terra
2026-07-09
Kimi K3
2026-07-16
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
- Kimi K3 has the higher public score estimate, 80.53 versus 72.94, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.136 vs $0.195. Cache-heavy agent loop: $0.2 vs $0.27.
- Context tradeoff
- Both models list 1.05M.
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 evidence55 rows
Agentic
Terminal-Bench 3.0
GPT-5.6 Terra20.8%SourceKimi K3—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Terra87.4%SourceKimi K388.3%SourceKimi K3 leads this result
BrowseComp
GPT-5.6 Terra87.5%SourceKimi K391.2%SourceKimi K3 leads this result
OSWorld 2.0
GPT-5.6 Terra50.2%SourceKimi K3—Not directly comparable
CyberGym
GPT-5.6 Terra81.8%SourceKimi K3—Not directly comparable
ExploitGym
GPT-5.6 Terra23.2%SourceKimi K3—Not directly comparable
Toolathlon
GPT-5.6 Terra53.1%SourceKimi K3—Not directly comparable
DeepSearchQA
GPT-5.6 Terra—Kimi K395.0%SourceNot directly comparable
Toolathlon-Verified
GPT-5.6 Terra—Kimi K373.2%SourceNot directly comparable
MCP Atlas
GPT-5.6 Terra—Kimi K384.2%SourceNot directly comparable
AutomationBench
GPT-5.6 Terra—Kimi K330.8%SourceNot directly comparable
JobBench
GPT-5.6 Terra—Kimi K352.9%SourceNot directly comparable
APEX-Agents
GPT-5.6 Terra—Kimi K337.6%SourceNot directly comparable
SpreadsheetBench 2
GPT-5.6 Terra—Kimi K334.8%SourceNot directly comparable
DECK-Bench
GPT-5.6 Terra—Kimi K373.5%SourceNot directly comparable
Coding
SWE-bench Pro
GPT-5.6 Terra63.4%SourceKimi K3—Not directly comparable
Terminal-Bench 2.0
GPT-5.6 Terra87.4%SourceKimi K3—Not directly comparable
deepSwe
GPT-5.6 Terra69.6%SourceKimi K367.5%SourceGPT-5.6 Terra leads this result
FrontierCode 1.1 Extended
GPT-5.6 Terra55.8%SourceKimi K3—Not directly comparable
cursorBench32
Shared sourceGPT-5.6 Terra64.9%Kimi K360.8%GPT-5.6 Terra leads this result
EEBench
Shared sourceGPT-5.6 Terra30.8%Kimi K338.3%Kimi K3 leads this result
FrontierSWE
GPT-5.6 Terra—Kimi K381.2%SourceNot directly comparable
ProgramBench
GPT-5.6 Terra—Kimi K377.8%SourceNot directly comparable
Kimi Code Bench v2
GPT-5.6 Terra—Kimi K372.9%SourceNot directly comparable
sweMarathon
GPT-5.6 Terra—Kimi K342%SourceNot directly comparable
PostTrain Bench
GPT-5.6 Terra—Kimi K336.6%SourceNot directly comparable
MLS-Bench Lite
GPT-5.6 Terra—Kimi K348.3%SourceNot directly comparable
VulcanBench v3
GPT-5.6 Terra—Kimi K373.9%SourceNot directly comparable
APEX-SWE
GPT-5.6 Terra—Kimi K348.0%SourceNot directly comparable
InferenceEval
GPT-5.6 Terra—Kimi K339.8%SourceNot directly comparable
KernelBench Internal
GPT-5.6 Terra—Kimi K368.5%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GPT-5.6 Terra92.9%SourceKimi K393.5%SourceKimi K3 leads this result
GPQA-D
GPT-5.6 Terra92.9%SourceKimi K393.5%SourceKimi K3 leads this result
HealthBench Professional
GPT-5.6 Terra57.7%SourceKimi K3—Not directly comparable
HealthBench Hard
GPT-5.6 Terra32.7%SourceKimi K3—Not directly comparable
HLE
GPT-5.6 Terra—Kimi K356%SourceNot directly comparable
HLE w/o tools
GPT-5.6 Terra—Kimi K343.5%SourceNot directly comparable
Math
Multimodal
MMMU-Pro
GPT-5.6 Terra80.7%SourceKimi K381.6%SourceKimi K3 leads this result
MMMU-Pro w/ Python
GPT-5.6 Terra82%SourceKimi K383.4%SourceKimi K3 leads this result
OfficeQA Pro
GPT-5.6 Terra—Kimi K363.3%SourceNot directly comparable
CharXiv w/o tools
GPT-5.6 Terra—Kimi K384.8%SourceNot directly comparable
CharXiv
GPT-5.6 Terra—Kimi K391.3%SourceNot directly comparable
MathVision
GPT-5.6 Terra—Kimi K394.3%SourceNot directly comparable
MathVision w/ Python
GPT-5.6 Terra—Kimi K397.8%SourceNot directly comparable
BabyVision w/ Python
GPT-5.6 Terra—Kimi K385.7%SourceNot directly comparable
ZeroBench
GPT-5.6 Terra—Kimi K323.0%SourceNot directly comparable
ZeroBench w/ Python
GPT-5.6 Terra—Kimi K341.0%SourceNot directly comparable
WorldVQA ForceAnswer
GPT-5.6 Terra—Kimi K351.0%SourceNot directly comparable
OmniDocBench
GPT-5.6 Terra—Kimi K391.1%SourceNot directly comparable
PerceptionBench
GPT-5.6 Terra—Kimi K358.5%SourceNot directly comparable
Frequently asked questions
Which is better, GPT-5.6 Terra or Kimi K3?
Kimi K3 has the higher public score estimate, 80.53 versus 72.94, 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-5.6 Terra or Kimi K3?
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, GPT-5.6 Terra or Kimi K3?
Kimi K3 leads the like-for-like agentic tasks comparison across 2 shared weighted benchmark rows.
Which costs less, GPT-5.6 Terra or Kimi K3?
For the stated presets, chat costs $0.008 on GPT-5.6 Terra and $0.0105 on Kimi K3; repository review costs $0.136 and $0.195; the cache-heavy agent loop costs $0.2 and $0.27. Costs use the listed standard API rates.
Which has the larger context window, GPT-5.6 Terra or Kimi K3?
Both models list the same context window, 1.05M.
Related comparisons
Last updated August 19, 2026
Watch GPT-5.6 Terra vs Kimi K3
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.