MoonshotAI
·Released on Apr 20, 2026Kimi K2.6 API Benchmarks, Pricing & Provider Data
Compare Kimi K2.6 with another model
Choose a model to open its comparison page.
LLM
Kimi K2.6 benchmark, API pricing, and provider data cover undefined API provider} other undefined API providers}}, with prices starting at $0.00005/request. Kimi K2.6 free API options are available from undefined provider} other undefined providers}}. The page also shows measured API speed and first-token latency.
Moonshot Kimi K2.6 is an open-source native multimodal agent model with 1 trillion MoE parameters, a 256K context window, and state-of-the-art coding, vision, and long-horizon agent swarm capabilities...
Quality
#48of 112
60.0
LMSpeed score
Speed
58char/s
18.93 s
Cost
#111of 186
$0.00005/ 1M · 8:1 in:out
$0.000008 in · $0.000042 out
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 7 / 8
- Methodology
- V3.0
#1Math58.4Estimated3/4 Measured dimensions
80% interval: 49.4–67.5
foundational mathPrior only
competition math59.9
aime · aime2026 / hmmt · hmmt_feb2026
proof frontier54.5
frontiermath · frontier_math_v2_tier4 / frontiermath · frontier_math_v2_tiers13
applied tool math60.9
mmanswer · mm_answer_bench
#2Reasoning55.3Estimated2/4 Measured dimensions
80% interval: 44.5–66.0
abstract logicPrior only
scientific causal53.9
critpt · critpt / gpqa · gpqa / hle · hle
multistep constraintsPrior only
evidence verification56.6
lcr · lcr
#3Instruction following54.8Provisional1/4 Measured dimensions
80% interval: 38.8–70.8
constraint followingPrior only
structured outputPrior only
novel instruction generalization54.8
ifbench · aa_if_bench
long multiturn instructionPrior only
#4Coding54.5RatedGlobal rank #174/4 Measured dimensions
80% interval: 48.4–60.5
code generation59.8
livecodebench · live_code_bench_v6 / scicode · scicode
repository engineering53
swe_pro · swe_pro / vibecode · vibe_code_bench
debugging testing55.7
swe_multilingual · benchlm_coding_swe_multilingual / swe_verified · swe_verified
tooling quality49.3
terminalbench · benchlm_coding_terminal_bench2
#5Agents53.6RatedGlobal rank #274/4 Measured dimensions
80% interval: 48.3–58.8
planning54
deep_planning · gert_labs
tool use50.9
mcp_atlas · mcp_atlas / tau · tau2_bench / toolathlon · toolathlon
environment execution55.4
browsecomp · browse_comp / deep_search_qa · deep_search_qa / gdpval_aa · benchlm_agentic_gdpval_aa / osworld · os_world2 / osworld · os_world_verified / terminalbench · benchlm_agentic_terminal_bench2 / wide_research · wide_research
recovery reliability54
apex_agents · apex_agents_aa / claw_eval · claw_eval / researchclaw · research_claw_bench
#6Multimodal53.4Estimated3/4 Measured dimensions
80% interval: 44.4–62.4
perception ocr61.3
v_star · v_star
document spatial47.9
charxiv · charxiv
visual reasoning51
mathvision · math_vision / mmmu_pro · mmmu_pro
video actionPrior only
#7Knowledge49.3Provisional1/4 Measured dimensions
80% interval: 35.3–63.3
broad knowledge49.3
aa_omniscience · aa_omniscience_index / benchlm_category_knowledge · benchlm_category_knowledge
professional knowledgePrior only
factualityPrior only
retrieval open bookPrior only
No data:Multilingual
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
INPUT
262.1Ktokens
≈ 314.6 pages of text
OUTPUT
235.9Ktokens
8K128K1M4M
262.1K
Features
Technical Details
- Input
- Output
- Released
- Apr 2026
- Tokenizer
- Other
- Architecture
- text+image->text
- Moderated
- No
- Supported parameters
- frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_pparallel_tool_callspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Rankings
Excels at
It's decent at
Falls behind in
Detailed scores
Updated: Sep 13, 2026Overall
undefined metric} other undefined metrics}}
Overall score60.0#48 / 112
Overall
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Pricing
undefined metric} other undefined metrics}}
Input price$0.950/M#111 / 186Output price$4.00/M#112 / 186
Pricing
undefined metric} other undefined metrics}}
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score53.6#2780% interval48.3–58.84/4 Measured dimensionsAgentic score61.0#41 / 77Terminal-Bench 2.066.7#21 / 53
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Planning & decomposition54
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
deep_planning · gert_labs · z 0.54 · q 1.00
Tool use50.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mcp_atlas · mcp_atlas · z -1.43 · q 1.00
tau · tau2_bench · z 0.74 · q 1.00
toolathlon · toolathlon · z 0.20 · q 1.00
Environment & long-horizon execution55.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
browsecomp · browse_comp · z 0.16 · q 1.00
deep_search_qa · deep_search_qa · z 0.81 · q 1.00
gdpval_aa · benchlm_agentic_gdpval_aa · z -0.13 · q 1.00
osworld · os_world2 · z -1.11 · q 1.00
osworld · os_world_verified · z 0.00 · q 1.00
terminalbench · benchlm_agentic_terminal_bench2 · z 0.19 · q 1.00
wide_research · wide_research · z 1.36 · q 1.00
Recovery & completion reliability54
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
apex_agents · apex_agents_aa · z 0.23 · q 1.00
claw_eval · claw_eval · z 0.45 · q 1.00
researchclaw · research_claw_bench · z 0.00 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Score54.5#1780% interval48.4–60.54/4 Measured dimensionsSciCode51.5%#28 / 89Coding score58.2#29 / 87
Coding
V3.0undefined metric} other undefined metrics}} · Rated
Dimensions and evidence
Code generation59.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
livecodebench · live_code_bench_v6 · z 1.21 · q 0.75
scicode · scicode · z 0.81 · q 1.00
Repository engineering53
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_pro · swe_pro · z 0.26 · q 1.00
vibecode · vibe_code_bench · z 0.47 · q 1.00
Debugging & testing55.7
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
swe_multilingual · benchlm_coding_swe_multilingual · z 0.16 · q 1.00
swe_verified · swe_verified · z 0.71 · q 1.00
Tool-assisted development & quality49.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
terminalbench · benchlm_coding_terminal_bench2 · z -0.07 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
SciCodeV3 evidence
LMSpeed rank#28
Score51.5%
Sourceartificialanalysis.ai
UpdatedSep 13, 2026
Confidence4
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Score55.380% interval44.5–66.02/4 Measured dimensionsGPQA91.1%#25 / 218HLE37.5%#36 / 216
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning53.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
critpt · critpt · z 0.39 · q 1.00
gpqa · gpqa · z 0.05 · q 1.00
hle · hle · z 0.22 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verification56.6
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
lcr · lcr · z 0.53 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Score49.380% interval35.3–63.31/4 Measured dimensionsKnowledge score56.6#63 / 83GPQA-D90.5#15 / 32
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Broad knowledge49.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aa_omniscience · aa_omniscience_index · z 0.30 · q 1.00
benchlm_category_knowledge · benchlm_category_knowledge · z -0.78 · q 1.00
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Artificial Analysis Intelligence IndexV3 evidence
LMSpeed rank#44
Score31.3
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence3
AA-Omniscience IndexV3 evidence
LMSpeed rank#30
Score5.3
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence3
AA-Omniscience Hallucination Rate
LMSpeed rank#86
Score40.5
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence3
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Score58.480% interval49.4–67.53/4 Measured dimensionsAIME2696.4#3 / 14HMMT Feb 202692.7#6 / 18
Math
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Foundational mathPrior only
Competition math59.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
aime · aime2026 · z 1.14 · q 1.00
hmmt · hmmt_feb2026 · z 0.97 · q 1.00
Advanced proofs54.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
frontiermath · frontier_math_v2_tier4 · z 0.49 · q 1.00
frontiermath · frontier_math_v2_tiers13 · z 0.41 · q 1.00
Applied & tool-assisted math60.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mmanswer · mm_answer_bench · z 1.50 · q 1.00
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
FrontierMath v2 (Tiers 1-3)V3 evidence
LMSpeed rank#12
Score39.0
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence3
FrontierMath v2 (Tier 4)V3 evidence
LMSpeed rank#16
Score14.6
Sourcebenchlm.ai
UpdatedSep 13, 2026
Confidence3
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only
Multimodal
V3.0undefined metric} other undefined metrics}} · Estimated
Score53.480% interval44.4–62.43/4 Measured dimensionsMultimodal Grounded score64.0#35 / 56MMMU-Pro79.4#12 / 31
Multimodal
V3.0undefined metric} other undefined metrics}} · Estimated
Dimensions and evidence
Perception & OCR61.3
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
v_star · v_star · z 1.25 · q 1.00
Document & spatial understanding47.9
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
charxiv · charxiv · z -0.13 · q 1.00
Visual reasoning51
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
mathvision · math_vision · z 0.00 · q 1.00
mmmu_pro · mmmu_pro · z 0.22 · q 1.00
Video & grounded actionPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Score54.880% interval38.8–70.81/4 Measured dimensionsAA-IFBench76.0#12 / 84
Instruction following
V3.0undefined metric} other undefined metrics}} · Provisional
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalization54.8
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
ifbench · aa_if_bench · z 0.35 · q 1.00
Multi-turn & long instructionsPrior only
Benchmark
LMSpeed rank
Score
Source
Updated
Confidence
OpenRouter endpoints
20 endpointsThird-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.
| Provider endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
DigitalOcean digitalocean | $0.950/M | $4/M | 100% | — | — | undefined tokens / undefined tokens |
SiliconFlow siliconflow/fp8 | $0.770/M | $3.40/M | 100.0% | — | — | undefined tokens / undefined tokens |
Cloudflare cloudflare | $0.950/M | $4/M | 99.9% | — | — | undefined tokens / undefined tokens |
Decart decart/fp4 | $0.587/M | $2.47/M | 99.8% | — | — | undefined tokens / undefined tokens |
Baidu baidu/fp4 | $0.580/M | $2.44/M | 99.7% | — | — | undefined tokens / undefined tokens |
CoreWeave coreweave/fp4 | $0.650/M | $3.41/M | 99.6% | — | — | undefined tokens / undefined tokens |
Moonshot AI moonshotai/int4 | $0.950/M | $4/M | 99.6% | — | — | undefined tokens / undefined tokens |
Parasail parasail/int4 | $0.750/M | $3.50/M | 99.6% | — | — | undefined tokens / undefined tokens |
Inceptron inceptron/int4 | $0.590/M | $2.45/M | 99.4% | — | — | undefined tokens / undefined tokens |
DeepInfra deepinfra/fp4 | $0.750/M | $3.50/M | 99.1% | — | — | undefined tokens / undefined tokens |
Novita novita | $0.800/M | $3.40/M | 98.6% | — | — | undefined tokens / undefined tokens |
AtlasCloud atlas-cloud/int4 | $0.950/M | $4/M | 97.5% | — | — | undefined tokens / undefined tokens |
StreamLake streamlake/fp8 | $0.598/M | $2.52/M | 97.1% | — | — | undefined tokens / undefined tokens |
Phala phala | $1.09/M | $4.60/M | 95.5% | — | — | undefined tokens / undefined tokens |
Venice venice/int4 | $0.750/M | $3.50/M | 94.4% | — | — | undefined tokens / undefined tokens |
Crusoe crusoe/bf16 | $0.700/M | $3.50/M | 93.1% | — | — | undefined tokens / undefined tokens |
Chutes chutes/int4 | $0.580/M | $3.40/M | 92.5% | — | — | undefined tokens / undefined tokens |
GMICloud gmicloud/fp8 | $0.855/M | $3.60/M | 90.7% | — | — | undefined tokens / undefined tokens |
BaseTen baseten/fp4 | $0.950/M | $4/M | 51.8% | — | — | undefined tokens / undefined tokens |
Fireworks fireworks | $0.950/M | $4/M | 0% | — | — | undefined tokens / undefined tokens |
Pricing Comparison
Compare Kimi K2.6 API pricing across 224 providers. Prices range from $0.00005/request to $100.00/request. CM-API 公益站 offers the lowest rate at $0.00005/request. 9 providers offer free API credits or a free tier.
| Provider | Health | Model Variant | Group | Input ($/M) | Output ($/M) | Speed (t/s) | First token | Audit |
|---|---|---|---|---|---|---|---|---|
L1 99% L2 9% | moonshot/kimi-k2.6 | default | -20%$0.760/M Cache read$0.128/M | -20%$3.20/M | 119.6 t/s | 19.53 s | ||
S3AI API Free | L1 94% L2 97% | Kimi-K2.6-free | free | Free | Free | 115.8 t/s | 18.26 s | — |
L1 94% L2 97% | Kimi-K2.6 | default | -29%$0.670/M Cache read$0.113/M | -30%$2.78/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -41%$0.557/M Cache read$0.094/M | -42%$2.31/M | 75.1 t/s | 14.12 s | — | |
L1 0% | kimi-k2.6 | 2api | $0.950/M Cache read$0.160/M | $4.00/M | 56.6 t/s | 9.32 s | ||
L1 0% | moonshotai/kimi-k2.6 | 懒人 | $0.950/M Cache read$0.200/M | $4.00/M | — | — | — | |
L1 100% | kimi-k2.6 | sale | $3.25/M | $13.50/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -53%$0.445/M | -54%$1.85/M | — | — | — | |
L1 100% | kimi-k2.6 | kimi-officially | -7%$0.882/M Cache read$0.149/M | -8%$3.66/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -6%$0.890/M Cache read$0.151/M | -8%$3.70/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -53%$0.445/M | -54%$1.85/M | — | — | — | |
L1 100% | KIMI-K2.6 | default | -6%$0.890/M | -7%$3.70/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -6%$0.890/M Cache read$0.150/M | -7%$3.70/M | — | — | — | |
L1 100% | Kimi-K2.6 | default | -6%$0.890/M | -7%$3.70/M | — | — | — | |
L1 100% | Pro/moonshotai/Kimi-K2.6 | default | $10.27/M | $10.27/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -25%$0.712/M | -26%$2.96/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -93%$0.065/M Cache read$0.011/M | -93%$0.274/M | — | — | — | |
L1 100% | kimi-k2.6 | Model-vip | -72%$0.267/M Cache read$0.045/M | -72%$1.11/M | — | — | — | |
L1 100% | kimi-k2.6 | default | -64%$0.342/M | -49%$2.05/M | — | — | — | |
L1 100% | kimi-k2.6 | deepseek | -63%$0.356/M Cache read$0.061/M | -63%$1.48/M | — | — | — |
Showing 20 model IDs of 111.
Alternatives & Similar Models
DeepSeek V4 Pro
deepseek-v4-pro
DeepSeek V4 Pro is the professional-tier DeepSeek V4 model, targeting frontier reasoning, coding, and agent workflows with maximum capability.
178 shared providers
DeepSeek V4 Flash
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
177 shared providers
GLM-5.1
glm-5-1
Zhipu GLM-5.1 is a next-generation GLM model aimed at frontier reasoning, coding, and bilingual agent applications.
168 shared providers
Kimi K2.5
kimi-k2-5
Moonshot Kimi K2.5 is an open-weight multimodal agent model with native vision and text input, strong coding performance, and a 256K context window.
161 shared providers
GPT-5.4
gpt-5-4
OpenAI GPT-5.4 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
155 shared providers
GLM-5.2
glm-5-2
Zhipu GLM-5.2 is Zhipu latest flagship coding and agentic model with a 1M-token context window, enhanced reasoning modes, and long-horizon software engineering capabilities.
155 shared providers
Frequently Asked Questions
- What benchmark data does Kimi K2.6 include?
- LMSpeed shows Kimi K2.6 benchmark context, API price, output speed, first-token latency, and provider data across 233 providers when those signals are available.
- What is the Kimi K2.6 API price?
- Kimi K2.6 has pricing from undefined provider} other undefined providers}}, ranging from $0.00005/request to $100.00/request. CM-API 公益站 has the lowest listed price.
- What does the Kimi K2.6 API pricing table include?
- The Kimi K2.6 API pricing table compares 233 providers by input price, output price, free tier status, speed, first-token latency, and recent health data when available.
- Which provider has the cheapest Kimi K2.6 API pricing?
- CM-API 公益站 currently has the lowest listed Kimi K2.6 price at $0.00005/request across undefined provider} other undefined providers}}.
- Can I compare Kimi K2.6 API price and speed together?
- Yes. LMSpeed shows Kimi K2.6 API price, output speed, first-token latency, and provider health on the same page so you can compare cost and performance together.
- Is Kimi K2.6 API free?
- Yes, Kimi K2.6 free API options are available through 9 providerundefined other undefined} on LMSpeed, including Zero API, 兔子API, DeadlySignal API, S3AI API, 兔子API. These providers offer free API credits or a free tier with no per-token charges.
- Where can I get Kimi K2.6 free API access?
- LMSpeed currently lists 9 free API providerundefined other undefined} for Kimi K2.6: Zero API, 兔子API, DeadlySignal API, S3AI API, 兔子API. Check each provider row before using it because free tier limits can change.