Model comparison

Kimi K2.6 vs Qwen3.8 Max

Updated August 3, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Kimi K2.6

Moonshot AI

58.5/100

Estimated · Public rank #65

90% interval 48.7–68.4

Qwen3.8 Max

Alibaba

65.4/100

Estimated · Public rank #31

90% interval 55.5–75.3

Qwen3.8 Max has the higher public score estimate, 65.4 versus 58.53, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

10 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
Qwen3.8 Max
Qwen3.8 Max has the larger documented context window.
Confidence: documented

Show secondary and unsupported calls

Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Cache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

22 Kimi K2.6 only

10 Shared

42 Qwen3.8 Max only

Shared results: 10
Kimi K2.6 only: 22
Qwen3.8 Max only: 42
Like-for-like categories: 2 / 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

Like-for-like

Kimi K2.6: 42.2
Qwen3.8 Max: 50.2
Weighted basis: 2 vs 2 rows
Reading: Qwen3.8 Max leads

Multimodal

Like-for-like

Kimi K2.6: 79.8
Qwen3.8 Max: 86.3
Weighted basis: 2 vs 2 rows
Reading: Qwen3.8 Max leads

Agentic

Directional only

Kimi K2.6: 73.5
Qwen3.8 Max: 86.1
Weighted basis: 3 vs 1 rows
Reading: Directional only

Coding

Directional only

Kimi K2.6: 64.4
Qwen3.8 Max: 67.7
Weighted basis: 3 vs 1 rows
Reading: Directional only

Reasoning

Not comparable

Kimi K2.6: Not measured
Qwen3.8 Max: 78.3
Weighted basis: 0 vs 2 rows
Reading: Not comparable

Math

Not comparable

Kimi K2.6: 67.1
Qwen3.8 Max: Not measured
Weighted basis: 4 vs 0 rows
Reading: Not comparable

Multilingual

Not comparable

Kimi K2.6: Not measured
Qwen3.8 Max: Not measured
Weighted basis: 0 vs 0 rows
Reading: Not comparable

Instruction following

Not comparable

Kimi K2.6: Not measured
Qwen3.8 Max: 82.8
Weighted basis: 0 vs 1 rows
Reading: Not comparable

Category averages with the server-provided evidence basis for Kimi K2.6 and Qwen3.8 Max
Category	Kimi K2.6	Qwen3.8 Max	Weighted basis	Reading
Knowledge	42.2	50.2	Like-for-like2 vs 2 rows	Qwen3.8 Max leads
Multimodal	79.8	86.3	Like-for-like2 vs 2 rows	Qwen3.8 Max leads
Agentic	73.5	86.1	Directional only3 vs 1 rows	Directional only
Coding	64.4	67.7	Directional only3 vs 1 rows	Directional only
Reasoning	Not measured	78.3	Not comparable0 vs 2 rows	Not comparable
Math	67.1	Not measured	Not comparable4 vs 0 rows	Not comparable
Multilingual	Not measured	Not measured	Not comparable0 vs 0 rows	Not comparable
Instruction following	Not measured	82.8	Not comparable0 vs 1 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.

CharXiv
Multimodal
Kimi K2.6: 80.4%Qwen3.8 Max: 93.5%Normalized gap 13.1Kimi K2.6 source Qwen3.8 Max source
OSWorld-Verified
Agentic
Kimi K2.6: 73.1%Qwen3.8 Max: 86.1%Normalized gap 13.0Kimi K2.6 source Qwen3.8 Max source
SWE-bench Pro
Coding
Kimi K2.6: 58.6%Qwen3.8 Max: 67.7%Normalized gap 9.1Kimi K2.6 source Qwen3.8 Max source
HLE
Knowledge
Kimi K2.6: 34.7%Qwen3.8 Max: 43.6%Normalized gap 8.9Kimi K2.6 source Qwen3.8 Max source
MMMU-Pro
Multimodal
Kimi K2.6: 79.4%Qwen3.8 Max: 82.3%Normalized gap 2.9Kimi K2.6 source Qwen3.8 Max source

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

Kimi K2.6: $0.00295; Fits in one request
Qwen3.8 Max: API rate not published; Fits in one request

Qwen3.8 Max has no comparable published API token rate.

Repository review

50K fresh input + 3K output tokens

Kimi K2.6: $0.0595; Fits in one request
Qwen3.8 Max: API rate not published; Fits in one request

Qwen3.8 Max has no comparable published API token rate.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Kimi K2.6: $0.249; Fits in one request; Cached input priced at the published list-input rate
Qwen3.8 Max: API rate not published; Fits in one request; Cached-input rate unavailable

Kimi K2.6 has no published cached-input rate, so cached tokens use its listed input rate. Qwen3.8 Max has no comparable published API token rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

SpecificationKimi K2.6Qwen3.8 Max

Context window

Maximum documented context; output-token limits may be lower.

Kimi K2.6

256K

Qwen3.8 Max

Alibaba Cloud Model Studio pricing

API model ID

Kimi K2.6

Not sourced

Qwen3.8 Max

qwen3.8-max

Alibaba Cloud Model Studio pricing

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Kimi K2.6

Not published

Qwen3.8 Max

No comparable hosted API rate

Alibaba Cloud Model Studio pricing

Documented inputs

Kimi K2.6

Not sourced

Qwen3.8 Max

Not sourced

Documented outputs

Kimi K2.6

Not sourced

Qwen3.8 Max

Not sourced

Provider availability

Kimi K2.6

Not sourced

Qwen3.8 Max

Not sourced

Reasoning profile

Kimi K2.6

Reasoning

Qwen3.8 Max

Reasoning

Weight access

Kimi K2.6

Open Weight

Qwen3.8 Max

Proprietary

License

Kimi K2.6

Open Weight

Qwen3.8 Max

Proprietary

Release date

Kimi K2.6

2026-04-20

Qwen3.8 Max

2026-08-03

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: Qwen3.8 Max has the higher public score estimate, 65.4 versus 58.53, but the 90% score intervals overlap.
Workload cost: A complete comparable API-rate estimate is not available for both models.
Context tradeoff: Qwen3.8 Max has the larger documented window (1M).

Run the same representative tasks against both endpoints before changing production traffic.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Kimi K2.6

API / mo$3,713

Self-host / mo$18,221

Break-even326M/day

Qwen3.8 Max

API / mo$0

Self-host / moNot listed

Break-even—

Proprietary model — self-hosting not applicable.

Model the full break-even

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence74 rows

Agentic

Terminal-Bench 2.0
Kimi K2.666.7%
Source
Qwen3.8 Max—
Not directly comparable
BrowseComp
Kimi K2.683.2%
Source
Qwen3.8 Max—
Not directly comparable
OSWorld-Verified
Kimi K2.673.1%
Source
Qwen3.8 Max86.1%
Source
Qwen3.8 Max leads this result
Toolathlon
Kimi K2.650%
Source
Qwen3.8 Max—
Not directly comparable
MCP Atlas
Kimi K2.655.9%
Source
Qwen3.8 Max—
Not directly comparable
Claw-Eval
Kimi K2.662.3%
Source
Qwen3.8 Max—
Not directly comparable
DeepSearchQA
Kimi K2.692.5%
Source
Qwen3.8 Max—
Not directly comparable
WideResearch
Kimi K2.680.8%
Source
Qwen3.8 Max81.9%
Source
Qwen3.8 Max leads this result
Gert Labs
Kimi K2.656.82%
Source
Qwen3.8 Max—
Not directly comparable
ResearchClawBench
Kimi K2.618.0%
Source
Qwen3.8 Max—
Not directly comparable
OSWorld 2.0
Kimi K2.64.6%
Source
Qwen3.8 Max19.4%
Source
Qwen3.8 Max leads this result
Terminal-Bench 2.1
Kimi K2.6—
Qwen3.8 Max86.6%
Source
Not directly comparable
CoWorkBench
Kimi K2.6—
Qwen3.8 Max74.8%
Source
Not directly comparable
JobBench
Kimi K2.6—
Qwen3.8 Max53.4%
Source
Not directly comparable
skillsBench
Kimi K2.6—
Qwen3.8 Max70.2%
Source
Not directly comparable
Agents' Last Exam
Kimi K2.6—
Qwen3.8 Max52.4%
Source
Not directly comparable
AutomationBench
Kimi K2.6—
Qwen3.8 Max27.3%
Source
Not directly comparable
Toolathlon-Verified
Kimi K2.6—
Qwen3.8 Max72.5%
Source
Not directly comparable
HLE w/ tools
Kimi K2.6—
Qwen3.8 Max56.2%
Source
Not directly comparable
WebArena-Verified
Kimi K2.6—
Qwen3.8 Max66.8%
Source
Not directly comparable
AndroidWorld
Kimi K2.6—
Qwen3.8 Max85.3%
Source
Not directly comparable
MobileWorld
Kimi K2.6—
Qwen3.8 Max77.8%
Source
Not directly comparable

Coding

SWE-bench Verified
Kimi K2.680.2%
Source
Qwen3.8 Max—
Not directly comparable
LiveCodeBench v6
Kimi K2.689.6%
Source
Qwen3.8 Max—
Not directly comparable
SWE-bench Pro
Kimi K2.658.6%
Source
Qwen3.8 Max67.7%
Source
Qwen3.8 Max leads this result
SWE Multilingual
Kimi K2.676.7%
Source
Qwen3.8 Max—
Not directly comparable
SciCode
Kimi K2.652.2%
Source
Qwen3.8 Max—
Not directly comparable
Terminal-Bench 2.0
Kimi K2.666.7%
Source
Qwen3.8 Max—
Not directly comparable
Vibe Code Bench
Kimi K2.637.89%
Source
Qwen3.8 Max—
Not directly comparable
cursorBench31
Kimi K2.647.6%
Source
Qwen3.8 Max—
Not directly comparable
Terminal-Bench 2.1
Kimi K2.6—
Qwen3.8 Max86.6%
Source
Not directly comparable
deepSwe
Kimi K2.6—
Qwen3.8 Max56.6%
Source
Not directly comparable
NL2Repo
Kimi K2.6—
Qwen3.8 Max55.9%
Source
Not directly comparable
FrontierSWE
Kimi K2.6—
Qwen3.8 Max73.5%
Source
Not directly comparable
MLS-Bench Lite
Kimi K2.6—
Qwen3.8 Max41.0%
Source
Not directly comparable
PaperBench
Kimi K2.6—
Qwen3.8 Max93.0%
Source
Not directly comparable

Reasoning

MRCRv2
Kimi K2.6—
Qwen3.8 Max92.9%
Source
Not directly comparable
LongBench v2
Kimi K2.6—
Qwen3.8 Max66.3%
Source
Not directly comparable

Knowledge

GPQA
Kimi K2.690.5%
Source
Qwen3.8 Max92.6%
Source
Qwen3.8 Max leads this result
GPQA-D
Kimi K2.690.5%
Source
Qwen3.8 Max92.6%
Source
Qwen3.8 Max leads this result
HLE
Kimi K2.634.7%
Source
Qwen3.8 Max43.6%
Source
Qwen3.8 Max leads this result
HLE w/o tools
Kimi K2.6—
Qwen3.8 Max43.6%
Source
Not directly comparable

Math

AIME26
Kimi K2.696.4%
Source
Qwen3.8 Max—
Not directly comparable
HMMT Feb 2026
Kimi K2.692.7%
Source
Qwen3.8 Max—
Not directly comparable
MMAnswerBench
Kimi K2.686.0%
Source
Qwen3.8 Max—
Not directly comparable
FrontierMath v2 (Tiers 1-3)
Kimi K2.638.966%
Source
Qwen3.8 Max—
Not directly comparable
FrontierMath v2 (Tier 4)
Kimi K2.614.580%
Source
Qwen3.8 Max—
Not directly comparable

Multimodal

MMMU-Pro
Kimi K2.679.4%
Source
Qwen3.8 Max82.3%
Source
Qwen3.8 Max leads this result
MMMU-Pro w/ Python
Kimi K2.680.1%
Source
Qwen3.8 Max—
Not directly comparable
CharXiv
Kimi K2.680.4%
Source
Qwen3.8 Max93.5%
Source
Qwen3.8 Max leads this result
MathVision
Kimi K2.687.4%
Source
Qwen3.8 Max95.2%
Source
Qwen3.8 Max leads this result
V*
Kimi K2.696.9%
Source
Qwen3.8 Max—
Not directly comparable
MathVision w/ Python
Kimi K2.6—
Qwen3.8 Max97.7%
Source
Not directly comparable
BabyVision
Kimi K2.6—
Qwen3.8 Max82.0%
Source
Not directly comparable
BabyVision w/ Python
Kimi K2.6—
Qwen3.8 Max91.3%
Source
Not directly comparable
ZeroBench
Kimi K2.6—
Qwen3.8 Max24.0%
Source
Not directly comparable
ZeroBench w/ Python
Kimi K2.6—
Qwen3.8 Max49.0%
Source
Not directly comparable
MedXpertQA (MM)
Kimi K2.6—
Qwen3.8 Max80.4%
Source
Not directly comparable
ScreenSpot Pro
Kimi K2.6—
Qwen3.8 Max84.5%
Source
Not directly comparable
Vision2Web
Kimi K2.6—
Qwen3.8 Max69.0%
Source
Not directly comparable
CharXiv w/o tools
Kimi K2.6—
Qwen3.8 Max88.4%
Source
Not directly comparable
OmniDocBench 1.5
Kimi K2.6—
Qwen3.8 Max92.1%
Source
Not directly comparable
OCRBench V2
Kimi K2.6—
Qwen3.8 Max74.2%
Source
Not directly comparable
CC-OCR
Kimi K2.6—
Qwen3.8 Max79.6%
Source
Not directly comparable
RealWorldQA
Kimi K2.6—
Qwen3.8 Max88.0%
Source
Not directly comparable
ERQA
Kimi K2.6—
Qwen3.8 Max77.8%
Source
Not directly comparable
SimpleVQA
Kimi K2.6—
Qwen3.8 Max75.0%
Source
Not directly comparable
PerceptionBench
Kimi K2.6—
Qwen3.8 Max63.5%
Source
Not directly comparable
Video-MME (with subtitle)
Kimi K2.6—
Qwen3.8 Max90.4%
Source
Not directly comparable
VideoMMMU
Kimi K2.6—
Qwen3.8 Max88.7%
Source
Not directly comparable
MMVU
Kimi K2.6—
Qwen3.8 Max82.4%
Source
Not directly comparable
MLVU (M-Avg)
Kimi K2.6—
Qwen3.8 Max90.8%
Source
Not directly comparable
LVBench
Kimi K2.6—
Qwen3.8 Max81.8%
Source
Not directly comparable

Instruction following

IFBench
Kimi K2.6—
Qwen3.8 Max82.8%
Source
Not directly comparable

Frequently asked questions

Which is better, Kimi K2.6 or Qwen3.8 Max?

Qwen3.8 Max has the higher public score estimate, 65.4 versus 58.53, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Kimi K2.6 or Qwen3.8 Max?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Kimi K2.6 or Qwen3.8 Max?

The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Kimi K2.6 or Qwen3.8 Max?

Both models do not have comparable published API token rates, so this page does not name a universal price winner.

Which has the larger context window, Kimi K2.6 or Qwen3.8 Max?

Qwen3.8 Max has the larger documented context window: 1M, compared with 256K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 3, 2026

Watch Kimi K2.6 vs Qwen3.8 Max

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.

Kimi K2.6 vs Qwen3.8 Max

Which one for your work

Long documents

Coding work

Agentic work

Chat turn cost

Cache-heavy agent loop cost

Repository review cost

What is actually comparable

Category results, on a stated basis

Knowledge

Multimodal

Agentic

Coding

Reasoning

Math

Multilingual

Instruction following

Shape of the matched evidence

What each workload costs

Chat turn

Repository review

Cache-heavy agent loop

Specification differences

Context window

API model ID

Cached-input rate

Documented inputs

Documented outputs

Provider availability

Reasoning profile

Weight access

License

Release date

Self-host vs API cost

Benchmark evidence

Agentic

Coding

Reasoning

Knowledge

Math

Multimodal

Instruction following

Frequently asked questions

Which is better, Kimi K2.6 or Qwen3.8 Max?

Which is better for coding, Kimi K2.6 or Qwen3.8 Max?

Which is better for agentic tasks, Kimi K2.6 or Qwen3.8 Max?

Which costs less, Kimi K2.6 or Qwen3.8 Max?

Which has the larger context window, Kimi K2.6 or Qwen3.8 Max?

Related comparisons

Watch Kimi K2.6 vs Qwen3.8 Max