ModelPriceWatch.com
Last scan 2026-10-08 Models tracked 278 Providers 37 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Kimi K2.5

by Fireworks

Retired budget open weights mid tier
Email me when Kimi K2.5 gets cheaper
Retired on . No longer served on the endpoint we price — the figures below are kept for historical reference only. Replaced by Kimi K2.6.
Availability: Retired as a per-token row: Fireworks no longer sells this model serverless. Its model page (fireworks.ai/models/fireworks/kimi-k2p5, read 2026-09-14) shows 'Serverless: Not supported', and the only way to buy it there is an On-demand Deployment on dedicated GPUs, billed per GPU-hour, so no per-token rate applies. Fireworks' size-tier fallback prices serverless calls only, so it does not apply either. Fireworks' rate card has not listed this model since at least 2026-08-09; it lists Kimi K2.6, K2.6 Fast, K2.7 Code, K2.7 Code Fast, K3, K3 Fast and K3 US instead. Fireworks publishes no delisting date, so the retirement date is the day we confirmed it, not a vendor notice. The $0.60/$3.00 per 1M (cached $0.10) shown is the last rate we published, and we never found it on a Fireworks page. The model page's meta description still quotes $0.60 uncached / $0.30 cached / $2.50 output, but that figure appears nowhere in the visible page, so we treat it as stale marketing copy rather than a price you can buy at. The release date is the one the model page gives for K2.5, 2026-01-27.

Today's price · per 1M tokens

Input

$0.600

per 1M tokens

Output

$3.00

per 1M tokens

Blended

$1.20

blended $/1M — 3:1 weighted input:output

Cached input

$0.100

17% of input — prompt caching

Source: official Fireworks pricing · read MODELPRICEWATCH.COM · 2026-10-08

Price receipt

Historical price.This model is retired — withdrawn on — so the provider no longer sells it and its page can no longer confirm this number: a price for a model no longer on sale is unconfirmable by definition, not missing. The figures above are the last prices we recorded for this model, kept as a historical record rather than a live quote.

Kimi K2.5 is still sold by Moonshot at $1.20 blended per 1M tokens — see the consolidated Kimi K2.5 page →

Overview

Kimi K2.5 on Fireworks. Withdrawn from serverless; last published rate $0.60/$3.00 per 1M, cached $0.10.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

RunPod →
Rent GPUs by the second and serve any open-weight model yourself.
Novita AI →
Serverless open-model inference billed per token — no GPU ops.
Vast.ai →
GPU marketplace — rent spot/on-demand GPUs cheaply and serve any open-weight model.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 2/4
Text ✓ Image ✓ Audio Video
Output
Text ✓
Features 4/9
Prompt caching ✓ Reasoning Coding Fast inference Long context ✓ Open weights ✓ Multimodal ✓ Web search Realtime

Benchmark performance

accuracy % · higher is better
Avg benchmark score62.9%
Perf per $/Mtok52.4
GPQA Diamond
87.6%
Terminal-Bench 2.1
50.8%
LiveCodeBench
85%
MCP Atlas
64.4%
Humanity's Last Exam
30.1%
ARC-AGI 2
12%
AIME 2025
96.1%
HumanEval
76.8%
AIME 2026 *
95.8%
Independent composite scores
  • AA Intelligence Index: 36 (Artificial Analysis composite across reasoning, knowledge and coding evals)
  • AA-Omniscience Index: -7.3 (−100–100)
How it stacks up
  • Ranks #63 of 98 benchmarked models by average score
  • Ranks #105 of 193 comparably-measured models by percentile score, across 10 independent measurements
  • Ranks #86 of 193 comparably-tested models by normalized performance per dollar
  • Strongest at AIME 2025 — 96.1%, #10 of 37

337.7 tokens/sec output

Source: Artificial Analysis, Vellum LLM Leaderboard, LMSYS Chatbot Arena (UC Berkeley) · updated · See full rankings →

Specifications

Provider
Fireworks
Context window
256K tokens
Modality
text, image
Parameters
Proprietary
Open source
Yes — open weights available
Released
Status
Retired ended
Last updated
Tags
open-weightsbudgetcachingmultimodal

Availability verified: — per Fireworks's own deprecation notice

Try on Fireworks Official pricing page