kimi-k3
Moonshot AI · text, image, video → text
Kimi K3 is Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window and industry-leading intelligence.
Input$3.00 /M
Output$15.00 /M
← Modelskimi-k3
muse-glimmer-30b
kimi-k3 vs muse-glimmer-30b
Output tokens cost $15.00 per million on kimi-k3 and $1.50 per million on muse-glimmer-30b. Input tokens cost $3.00 per million on kimi-k3 and $0.35 per million on muse-glimmer-30b. Cached input tokens are billed at $0.30 per million on kimi-k3 and $0.04 per million on muse-glimmer-30b. Context lengths are 1,048,576 tokens on kimi-k3 and 131,000 tokens on muse-glimmer-30b. Time to first token (TTFT) measured on AIHubMix is 3.5s on kimi-k3 and 0.4s on muse-glimmer-30b. Measured output throughput is 27.1 tok/s on kimi-k3 and 42.5 tok/s on muse-glimmer-30b. On the LMArena coding leaderboard kimi-k3 scores 1531 and muse-glimmer-30b scores 1482.
kimi-k3muse-glimmer-30b+ Add model · 2/4
muse-glimmer-30b
Meta · text, image → text
Muse Glimmer 30B is a dense causal language model distilled from Muse Spark, designed for autonomous agent work. It combines multi-step reasoning, robust pattern-based tool invocation, and fault-recovery capabilities, and offers multimodal understanding via a ViT-G/14 perceptual encoder of roughly 1.8 billion parameters, supporting interleaved text and image inputs, a context window of over 131K tokens, and optional inference strength settings (from low to ultra-high). Muse Glimmer is trained on data in over 100 languages, performs strongly for its size on agent benchmarks including MCP Atlas, DeepSearch QA, Gaia2, and SWE-Bench Pro, and is released under the Apache 2.0 license.
Input$0.35 /M
Output$1.50 /M
Pricing & Specifications
Prices are per million tokens. Time to First Token and throughput are rolling averages measured on AIHubMix.
kimi-k3
muse-glimmer-30b
Input /M
$3.00
$0.35
Output /M
$15.00
$1.50
Cache read /M
$0.30
$0.04
Context length
1,048,576
131,000
Max output
1,048,576
131,000
Time to First Token
3.5 s
0.4 s
Throughput
27.1 tok/s
42.5 tok/s
Modalities
textimagevideo
textimage
Supported Parameters
thinkingfunction callingstructured outputs
thinkingtools
API Formats
Promotional prices show the discounted rate; see each model page for promotion windows.
Activity Past 30 Days
Daily traffic served through AIHubMix — how demand for each model is trending.
kimi-k3muse-glimmer-30b
Tokens / day
-
Requests / day
-
Performance Past 3 Days
Measured on real AIHubMix traffic, hourly buckets. Gaps mean no traffic in that hour.
kimi-k3muse-glimmer-30b
Throughput (tok/s)
-
TTFT (s)
-
Uptime (%)
-
LMArena Benchmarks
LMArena ratings by capability (Bradley-Terry, commonly called Elo). Higher is better.
Text
kimi-k3muse-glimmer-30b
1340140014601520
Overall
14261486
Coding
14821531
Math
14531453
Hard prompts
14451505
Instruction following
14141479
Multi-turn
14351499
Creative writing
13721470
Longer query
14211502
Chinese
14801527
English
14261491
Source: LMArena (arena.ai) leaderboard, imported by AIHubMix. Models without published ratings are omitted per chart.
Cost calculator
Estimate your monthly bill for the same workload on each model.
muse-glimmer-30b
$43.50 /mo
kimi-k3
$405 /mo
Monthly = daily × 30. Discounted rates applied where a promotion is active.
FAQ
Which is cheaper: kimi-k3, muse-glimmer-30b?
muse-glimmer-30b: $1.50/M output tokens; kimi-k3: $15.00/M. Use the cost calculator above to estimate your own workload.
How do their coding arena scores compare?
kimi-k3: 1531; muse-glimmer-30b: 1482 (LMArena coding leaderboard).
Which responds faster?
muse-glimmer-30b: 0.4s time to first token measured on AIHubMix; see the live performance charts above for how each model behaves across the day.
How large is each context window?
kimi-k3 accepts 1,048,576 and muse-glimmer-30b accepts 131,000 input tokens. Maximum output per request is 1,048,576 tokens on kimi-k3 and 131,000 tokens on muse-glimmer-30b.
Which one generates tokens faster?
muse-glimmer-30b at 42.5 tok/s and kimi-k3 at 27.1 tok/s, measured as output throughput on AIHubMix — a separate metric from time to first token.
What inputs and capabilities does each model support?
kimi-k3 accepts text, image and video input and supports thinking, function calling and structured outputs; muse-glimmer-30b accepts text and image input and supports thinking and tool calling.
Can I call kimi-k3 and muse-glimmer-30b with the same API key?
Yes. AIHubMix serves every model on this page behind one OpenAI-compatible endpoint, so switching between them is a one-line change to the model field — no second account, key or SDK.
Popular comparisons
Related model match-ups readers also look at.
AIHubMix