MoonshotAI: Kimi K3 (batch)

moonshotai/kimi-k3-batch

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI.

Modalities
In / out per 1M$2.28 / $11
Context1M
ReleasedJul 16, 2026

Providers

25 providers serve this model upstream.

Providers serving this model
ProviderInput /MOutput /MCache read /MLatency P50Throughput P50Uptime
Relacefp4$0.54$14$0.301.81s51 tps100%
DigitalOcean$2.55$13$0.261.07s40 tps100%
Moonshot AImxfp4$3.00$15$0.305.44s24 tps100%
inference.netfp4$0.70$15$0.321.05s77 tps99.97%
Waferus$2.80$16$0.300.72s52 tps99.97%
inference.net Fastfp4$2.99$18$0.450.89s94 tps99.97%
Wafer$0.72$24$0.300.93s51 tps99.95%
Fireworks Fast$4.50$23$0.451.36s70 tps99.95%
Parasailfp4$2.60$13$0.261.45s72 tps99.92%
Fireworksus$4.50$23$0.451.28s42 tps99.9%
Phala$2.55$13$0.261.53s72 tps99.89%
Modalmxfp4$3.00$15$0.301.90s60 tps99.82%
Alibaba Cloud Int.$3.45$17$0.343.66s33 tps99.79%
Basetenfp8$3.00$15$0.303.76s77 tps99.75%
Fireworks$3.00$15$0.301.24s28 tps99.71%
DeepInfraus ยท mxfp4$2.85$14$0.280.96s40 tps98.81%
Parasail Fastfp4$3.00$15$0.300.55s94 tps98.68%
Together$3.00$15$0.300.96s55 tps98.33%
Chutesmxfp4$1.85$9.50$0.183.24s17 tps96.17%
Amazon Bedrockus-east-2$3.00$15$0.304.27s52 tps95.27%
Makorafp4$1.53$13$0.200.53s128 tps94.62%
AkashMLfp4$1.20$14$1.201.11s76 tps92.31%
Morphfp8$0.67$20$0.293.01s26 tps89.95%
Sail Researchfp4$0.84$14$0.302.22s70 tps89.7%
Decartmxfp4$2.31$12$0.231.96s71 tps83.93%

Market-wide measurements, refreshed every 15 minutes. They describe who serves Kimi K3 (batch) in general, not how ProjectCOZY routes your requests.

Pricing

List rates per 1M tokens โ€” what Kimi K3 (batch) costs pay-as-you-go elsewhere, not what you pay here.

Input
$2.28
per 1M tokens
Output
$11
per 1M tokens
Cache read
$0.23
per 1M cached tokens
Included in every pass
Kimi K3 (batch) is included in every ProjectCOZY pass at no extra cost โ€” one flat rate, no per-token metering. The rates above are there for comparison.

Performance

Time to first token and output speed across every provider, over the last 3 days.

0.0s200.0s400.0s600.0s800.0s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
Together Alibaba Cloud Int. inference.net Fast Decart Morph Sail Research Moonshot AI AkashML Amazon Bedrock (us-east-2) Fireworks (us) Fireworks Fast Wafer (us) Relace DeepInfra (us) Phala Parasail Fast Makora inference.net Wafer Modal Baseten DigitalOcean Parasail Fireworks Chutes
0 tok/s500 tok/s1000 tok/s1500 tok/s2000 tok/s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
Together Alibaba Cloud Int. inference.net Fast Decart Morph Sail Research Moonshot AI AkashML Amazon Bedrock (us-east-2) Fireworks (us) Fireworks Fast Wafer (us) Relace DeepInfra (us) Phala Parasail Fast Makora inference.net Wafer Modal Baseten DigitalOcean Parasail Fireworks Chutes

Uptime

100% of requests were served successfully over the last 3 days, across all providers.

Last 3 daysAvailability 99.58%
Oct 11, 2:00 PMNow
Availability over the last 24 hoursAvailability 99.85%
100% 98% 97% 95%
Oct 13, 2:40 PMNow

Apps

Tools that most often run Kimi K3 (batch), by tokens this week.

Activity

7790.4B tokens processed market-wide in the last month.

0200.0B400.0B600.0B800.0B
Sep 14Sep 19Sep 24Sep 29Oct 4Oct 9Oct 14
Tokens

FAQ

Which model id do I use for Kimi K3 (batch)?
moonshotai/kimi-k3:batch. Send it as the model field from any OpenAI-compatible client, against http://192.168.1.55/v1.
How much context does Kimi K3 (batch) hold?
1,048,576 tokens per request, of which up to 16,384 can be the response.
What can it accept and return?
Input: text, image, video. Output: text.
What does Kimi K3 (batch) cost on a pass?
Nothing beyond the pass itself. Pay-as-you-go elsewhere it lists at $2.28 per 1M input tokens and $11 per 1M output tokens.
Who built Kimi K3 (batch)?
MoonshotAI, in the Kimi family.

Explore

Other models your pass can call.