Kimi K2 Instruct 0905

Kimi K2 Instruct 0905 on Zumik: live pricing, context, and caching, routable by id or alias through one OpenAI-compatible endpoint.

$0.57
Input / 1M tokens
$2.30
Output / 1M tokens
$0.57
Cache read
131K
Context window

At a glance.

ProviderFireworks AI
Familydeepseek_v3
Released2025-07
LicenseOpen weights
Context window131K tokens
Max output33K tokens
Parameters1028B
Modalitiestext
Tool callingYes
Reasoning modeNo
Cachingnone
Batch discountNo batch tier

What reuse looks like here.

Not yet profiled

Pricing, context, and capabilities for Kimi K2 Instruct 0905 are live, but it is outside the flagship set Zumik benchmarks in depth, so measured reuse, capture, and warm TTFT are not shown yet. Run a workload estimate or route it by id to start collecting traces.

Reuse economics

What you actually pay once caching works.

At a typical 55% prefix reuse, a million input tokens on Kimi K2 Instruct 0905 effectively costs $0.57 instead of $0.57 - blending to roughly $1.00 with a 25% output share. There is no batch tier, so cost control here leans on caching and routing.

Estimate it for your workload
Best for
text

Route it directly by id, or let an alias pick it when it wins under policy.

Same OpenAI client, this model.

python
from openai import OpenAI

client = OpenAI(base_url="https://api.zumik.ai/v1", api_key="zk_live_...")

r = client.responses.create(
    model="kimi-k2-instruct-0905",
    input="Draft a fix for the failing test.",
)
print(r.usage.input_tokens_cached)   # confirm reuse

Kimi K2 Instruct 0905, answered.

How much does Kimi K2 Instruct 0905 cost?

Kimi K2 Instruct 0905 is $0.57 per million input tokens and $2.30 per million output tokens through Zumik. Cache reads are $0.57 per million, a 0% discount on input.

What is Kimi K2 Instruct 0905's context window?

Kimi K2 Instruct 0905 supports a 131K-token context window with up to 33K output tokens.

Run Kimi K2 Instruct 0905 with reuse measured.

Point an OpenAI client at Zumik and see exactly how much of this model's input you are reusing.