TL;DR. Artificial Analysis positions Kimi K3's overall intelligence level alongside GPT-5.5 and Claude Opus 4.8. On the GDPval v2 agentic evaluation, K3 (1668 Elo) actually scores above both Opus 4.8 (1600) and GPT-5.5 (1494). Only Claude Fable 5 and GPT-5.6 Sol rank clearly higher. K3 achieves this at a lower price: moonshotai/kimi-k3 runs $3/$15 per million on ofox compared to $5/$30 for GPT-5.5 and $5/$25 for Opus 4.8. On cost-per-task measures, K3 comes in around $0.94, roughly half of Opus 4.8's $1.80. For frontier-level work, K3 is the value selection that costs less per token and approximately half per finished task compared to Opus 4.8.
TL;DR: Which One Should You Pick?
One-line verdict: K3 represents the value selection against GPT-5.5 and Opus 4.8, matching them on intelligence while outperforming them on agentic tasks at a lower cost. Only upgrade to the absolute top tier (GPT-5.6 Sol, Fable 5) or for specific ecosystem requirements.
The key reframing: K3 does not chase the top position. It aligns with the previous cycle's frontier (GPT-5.5) and the current Claude flagship (Opus 4.8) while charging less than either.
| Your priority | Pick | Why |
|---|---|---|
| Agentic coding and tool use | Kimi K3 | Tops Opus 4.8 and GPT-5.5 on AA's GDPval v2 agentic Elo, at a lower price |
| Lowest cost at frontier-adjacent quality | Kimi K3 | ~$0.94 per task on AA vs Opus 4.8's $1.80, comparable intelligence |
| Deep in the Anthropic ecosystem | Claude Opus 4.8 | Comparable intelligence to K3; you pay for the ecosystem, not a capability gap |
| A known, broad OpenAI generalist | GPT-5.5 | Solid all-rounder, though K3 matches its intelligence and undercuts its price |
| The outright top of the board | GPT-5.6 Sol or Fable 5 | These are the models K3 still trails; buy them when the ceiling changes the outcome |
| Budget, text-only coding | Kimi K2.7 Code | A third of K3's price; covered in its own section below |
For agentic and coding work at the level GPT-5.5 and Opus 4.8 operate, K3 is both the strongest on the relevant benchmark and the least expensive.
flowchart TD
A[Task needs frontier-ish quality] --> B{Need the absolute top of the board?}
B -->|Yes| C[GPT-5.6 Sol or Claude Fable 5]
B -->|No| D{Locked into the Anthropic ecosystem?}
D -->|Yes| E[Claude Opus 4.8<br/>anthropic/claude-opus-4.8]
D -->|No| F{Budget-first, text-only coding?}
F -->|Yes| G[Kimi K2.7 Code<br/>moonshotai/kimi-k2.7-code]
F -->|No| H[Kimi K3<br/>moonshotai/kimi-k3<br/>peer intelligence, lower cost]
Quick Specs Comparison
Prices verified on the ofox catalog, July 17 2026, per million tokens. All three flagships accept image input; K3's context window (1M) is its standout.
| Spec | Kimi K3 | GPT-5.5 | Claude Opus 4.8 | Kimi K2.7 Code |
|---|---|---|---|---|
| ofox model ID | moonshotai/kimi-k3 |
openai/gpt-5.5 |
anthropic/claude-opus-4.8 |
moonshotai/kimi-k2.7-code |
| Input /M | $3.00 | $5.00 | $5.00 | $0.95 |
| Output /M | $15.00 | $30.00 | $25.00 | $4.00 |
| Cache read /M | $0.30 | $0.50 | $0.50 | $0.19 |
| Image input | Yes | Yes | Yes | No |
| AA Intelligence Index | 57 | comparable to K3 | comparable to K3 | n/a |
K3 is the least expensive of the three frontier-class models on every pricing line—input, output, and cache—while Artificial Analysis rates its intelligence level with the other two. K2.7 Code appears in the table as the budget reference: substantially cheaper, but text-only and a smaller model, representing a different tier of decision.
What Kimi K3 Actually Is
Moonshot describes K3 as the first open 3T-class model. The architecture is a 2.8 trillion parameter Mixture-of-Experts they call Stable LatentMoE, activating 16 of 896 experts per token, with an attention stack that pairs Kimi Delta Attention (KDA) with an AttnRes residual scheme. This makes it substantially larger than competing open-weight models (GLM-5.2 is 753B, DeepSeek V4 Pro 1.6T). The practical highlights are simpler: 1M token context, native image input, and integrated thinking.
The thinking operates similarly to the rest of the current Kimi line, with reasoning exposed in a reasoning_content field. At launch K3 runs at maximum thinking effort by default, and Moonshot indicates low- and high-effort modes are coming in subsequent updates, so presently the effort tier is not adjustable. Reasoning tokens bill as output at $15/M, so a verbose maximum-thinking run costs more. Token efficiency remains strong—K3 used approximately 21% fewer output tokens than K2.6 to complete the Artificial Analysis index while scoring higher.
On ofox, K3 is moonshotai/kimi-k3 on the OpenAI-compatible endpoint, and the OpenAI protocol path works directly. If accessed through the Anthropic-compatible path instead, the thinking parameter is not forced, which differs from K2.7 behavior. The open weights have not yet released; Moonshot commits to release by July 27 2026, and the license was not announced in the launch post. For now, K3 is an API model.
How K3 Compares to GPT-5.5 and Opus 4.8: Three Reads
Cross-model comparison is where most analyses mix numbers from different labs run on different systems. This section maintains each evaluation to a single source for honest comparison.
Read 1: Overall intelligence (Artificial Analysis Index)
The Artificial Analysis Intelligence Index combines numerous benchmarks into one number so models from different vendors appear on the same scale. K3 scores 57. Artificial Analysis characterizes that intelligence as comparable to Claude Opus 4.8 and GPT-5.5, and behind Claude Fable 5 and GPT-5.6 Sol, which lead the board near 60 and 59. This is a rolling snapshot taken around K3's July 2026 launch, so read it as an ordering with a date, not a fixed score. K3 has reached parity with GPT-5.5 and Opus 4.8 on general intelligence, and has not caught the very top tier.
Read 2: Agentic tasks (AA GDPval v2)
Overall intelligence understates K3 on the work people actually automate. On Artificial Analysis's GDPval v2 agentic evaluation, all run on the same harness, K3 ranks above both Opus 4.8 and GPT-5.5, with only Claude Fable 5 and GPT-5.6 Sol higher.
| Model | GDPval v2 Elo (Artificial Analysis) |
|---|---|
| Claude Fable 5 | 1760 |
| GPT-5.6 Sol (max) | 1748 |
| Kimi K3 | 1668 |
| Claude Opus 4.8 | 1600 |
| GLM-5.2 | 1514 |
| GPT-5.5 | 1494 |
| Kimi K2.6 | 1190 |
This is a substantive result: for agentic tool-use and multi-step work, K3 out-benchmarks the two comparison models. GDPval v2 scores models on realistic, economically valuable tasks rather than trivia, so a lead here maps to the kind of work a coding or ops agent actually performs. K3 also takes the top spot on AA's AutomationBench (their implementation of Zapier's agentic SaaS-workflow evaluation) and reaches 1547 on AA-Briefcase, a private long-horizon knowledge-work eval, second only to Fable 5 and up 732 points from Kimi K2.6. Across three separate agentic tests, the pattern remains consistent: K3 is at or near the top of each, and by far the least expensive of the models scoring at that level.
Why does an open-ish Kimi lead two closed frontier models on agentic work while sitting level with them on the general index? Because agentic evals reward planning, tool use, and sustained focus across many steps, and K3's default maximum-thinking budget plus its token efficiency are tuned for exactly that. GPT-5.5 and Opus 4.8 are strong generalists, but neither was designed to top an autonomous-agent leaderboard the way this Kimi generation was. For a single-shot prompt, the general index (where all three are comparable) is the authoritative read. For an agent that runs for many turns, GDPval is the relevant measure, and K3 wins it.
Read 3: Cost per task (Artificial Analysis)
Per-token prices slightly favor K3, because a model that reasons more can burn more tokens per job. The fairer money read is cost per task, which Artificial Analysis measures across its whole index.
| Model | Cost per task (Artificial Analysis) |
|---|---|
| DeepSeek V4 Pro | $0.04 |
| GLM-5.2 | $0.32 |
| Kimi K3 | $0.94 |
| GPT-5.5 (xhigh) | $0.99 |
| GPT-5.6 Sol | $1.04 |
| Claude Opus 4.8 | $1.80 |
K3 lands around $0.94 per task, approximately even with GPT-5.5's $0.99 (from AA's June v4.1 snapshot) and GPT-5.6 Sol's $1.04, and roughly half of Opus 4.8's $1.80. Note the two money reads diverge for the GPT-5.x line: per token K3 is clearly cheaper ($3/$15 vs $5/$30), but per task it lands about even with GPT-5.5, because GPT-5.5 puts fewer tokens through each job. Against Opus 4.8, K3 is cheaper on both reads. K3 is materially cheaper than Opus 4.8 for comparable intelligence, cheaper per token than the GPT-5.x flagships, and roughly tied with GPT-5.5 once token usage is counted.
Do not cross-reference K3's Moonshot GPQA score against a GPT-5.5 or Opus figure from another system and call it a head-to-head. The AA reads are the ones where these models sit on a comparable scale.
When to Pick Each One
Against GPT-5.5 and Opus 4.8, the decision comes down to capability-per-dollar plus ecosystem:
Pick Kimi K3 for agentic and coding work at frontier-adjacent quality. It matches GPT-5.5 and Opus 4.8 on overall intelligence, beats them on AA's agentic eval, and costs less than both, with a 1M context, vision, and open weights on the way. It is the default value choice of the three.
Pick Claude Opus 4.8 when you are already in the Anthropic ecosystem or want its specific behavior and tooling. Artificial Analysis rates its intelligence level with K3 while its cost per task runs about double, so you are paying for the ecosystem, not a capability gap.
Pick GPT-5.5 when you want OpenAI's broad, well-understood generalist and the surrounding tooling. It is a strong all-rounder, but K3 matches its intelligence, beats its agentic Elo, and undercuts its price, so the reason to pick it is familiarity and stack, not raw value.
Step up to GPT-5.6 Sol or Fable 5 when the task genuinely needs the top of the board. These are the models K3 still trails on the intelligence index. Hard reasoning where the last couple of points change the outcome justifies the extra cost. Outside coverage frames K3 the same way: The Decoder reads it as nearing the closed leaders rather than passing them.
If K3 is off your table for policy or preference and the choice is just GPT-5.5 versus Opus 4.8, the two are close on the general index but split by character. On AA's GDPval v2 agentic eval, Opus 4.8 (1600) sits well above GPT-5.5 (1494), so for tool use and multi-step work Opus is the stronger of the pair. On price they match on input ($5/M each) with Opus a little cheaper on output ($25 vs $30/M). GPT-5.5's advantage is the breadth and maturity of OpenAI's tooling and the familiarity of its behavior in existing stacks. Between those two: Opus 4.8 for agentic and reasoning-heavy work, GPT-5.5 for a broad, well-supported generalist. K3 leads this comparison because it matches both on intelligence and beats both on the agentic eval while costing less than either.
Monthly bill: one developer, agentic coding
To make the price gap concrete, here is a per-token monthly bill for a single developer running a reasoning-heavy agent at 20M input and 5M output tokens a month.
| Monthly workload (20M in / 5M out) | Kimi K3 | GPT-5.5 | Claude Opus 4.8 |
|---|---|---|---|
| Per-token bill | $135 | $250 | $225 |
K3 runs about 54% of GPT-5.5's bill and 60% of Opus 4.8's on the same traffic. Scale that across a five-developer team and K3 saves roughly $575 a month against GPT-5.5 and $450 against Opus 4.8, for work Artificial Analysis rates as comparable in intelligence and stronger on agentic tasks.
The Cheaper Kimi: K2.7 Code
If your work is plain text coding and the frontier is overkill, the relevant comparison is not GPT-5.5 or Opus 4.8, it is a smaller Kimi. Kimi K2.7 Code (moonshotai/kimi-k2.7-code) is a 1T-total, 32B-active, text-only MoE tuned for code, at $0.95/$4 per million with a 256K context and a Modified MIT open-weight license. It is about a third of K3's price.
| Monthly workload (1 developer) | K2.7 Code | Kimi K3 | K3 premium |
|---|---|---|---|
| 20M input / 5M output, no cache | $39.00 | $135.00 | 3.5× |
| Vision task: 10M input / 3M output | not possible (text-only) | $75.00 | n/a |
The rule is straightforward. If the job is text-only coding on a budget, K2.7 Code does it for a fraction of the price and the frontier headroom sits unused. If the job needs vision, a 1M context, or the agentic strength in the benchmarks above, K2.7 Code cannot reach it and you are back to K3.
Where K3 Lands on Moonshot's Own Benchmarks
For completeness, here are the vendor numbers. Every figure is Moonshot-reported, run at maximum thinking, from the official K3 launch post. Treat them as direction, not independently reproduced, and do not compare them cell-by-cell against the AA reads above, which use different harnesses.
| Benchmark | Kimi K3 (max thinking), Moonshot-reported |
|---|---|
| DeepSWE | 67.5 |
| Terminal Bench 2.1 | 88.3 |
| Program Bench | 77.8 |
| GPQA-Diamond | 93.5 |
| MathVision (with Python) | 97.8 |
| BrowseComp | 91.2 |
Those track with the third-party picture: strong on science reasoning (GPQA-Diamond) and web-research agents (BrowseComp), consistent with the agentic strength AA measured. Moonshot's own post is candid that K3 still trails Claude Fable 5 and GPT-5.6 Sol overall, which is why this article positions it as a cheaper peer of GPT-5.5 and Opus 4.8 rather than a new number one.
Run All Three on ofox: A/B in a Few Lines
K3, GPT-5.5, and Opus 4.8 all live on the same OpenAI-compatible endpoint, so testing them against your own task is a one-string swap. Point the SDK at https://api.ofox.ai/v1, loop over the three model IDs, and read usage to compare real token cost.
Python: A/B all three in one loop
from openai import OpenAI
client = OpenAI(base_url="https://api.ofox.ai/v1", api_key="YOUR_OFOX_KEY")
task = "Refactor this function to be async and add error handling:\n\n"
task += open("handler.py").read()
for model in ["moonshotai/kimi-k3", "openai/gpt-5.5", "anthropic/claude-opus-4.8"]:
r = client.chat.completions.create(
model=model,
messages=[{"role": "user", "content": task}],
)
print(f"\n=== {model} ===")
print(r.usage) # compare tokens, then multiply by the specs-table prices
print(r.choices[0].message.content)
Node: same shape
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.ofox.ai/v1",
apiKey: process.env.OFOX_API_KEY,
});
const task = "Refactor this function to be async and add error handling:\n" + code;
for (const model of ["moonshotai/kimi-k3", "openai/gpt-5.5", "anthropic/claude-opus-4.8"]) {
const r = await client.chat.completions.create({
model,
messages: [{ role: "user", content: task }],
});
console.log(`\n=== ${model} ===`);
console.log(r.usage);
console.log(r.choices[0].message.content);
}
Attach a screenshot (all three take images)
All three flagships accept image input, so a vision task runs on any of them by swapping the model string. Send the image as an image_url block. The one model this fails on is moonshotai/kimi-k2.7-code, which is text-only.
import base64
with open("layout-bug.png", "rb") as f:
b64 = base64.b64encode(f.read()).decode()
r = client.chat.completions.create(
model="moonshotai/kimi-k3",
messages=[{
"role": "user",
"content": [
{"type": "text", "text": "This UI screenshot has a layout bug. What is wrong and how do I fix the CSS?"},
{"type": "image_url", "image_url": {"url": f"data:image/png;base64,{b64}"}},
],
}],
)
print(r.choices[0].message.content)
Run the first loop on a representative slice of your real tasks, sum the usage across a day, and multiply by the specs-table prices. That gives you the actual K3-vs-GPT-5.5-vs-Opus delta for your traffic, which beats any leaderboard for your specific workload.
FAQ
Is Kimi K3 better than GPT-5.5? On Artificial Analysis, K3's overall intelligence (57) is rated comparable to GPT-5.5, and on the GDPval v2 agentic eval K3 (1668) scores above GPT-5.5 (1494). K3 is also cheaper: $3/$15 vs $5/$30. GPT-5.5 remains a strong generalist, but on value K3 wins for agentic and coding work.
Is Kimi K3 better than Claude Opus 4.8? Artificial Analysis rates their overall intelligence comparable (K3 scores 57, a rolling snapshot), and K3 (1668) edges Opus 4.8 (1600) on GDPval v2. K3 is cheaper, $3/$15 vs $5/$25 per token and about half the cost per task ($0.94 vs $1.80 on AA's runs). Choose Opus 4.8 for the Anthropic ecosystem, K3 for similar intelligence at roughly half the cost.
How much does Kimi K3 cost on ofox? $3/M input, $15/M output, $0.30/M cache read. That is below both GPT-5.5 and Opus 4.8, which are $5/M input each. The prices match Moonshot's own API, so there is no ofox markup.
Which is best for agentic coding, K3, GPT-5.5, or Opus 4.8? Among these three, AA's GDPval v2 agentic eval orders K3 (1668) ahead of Opus 4.8 (1600) and GPT-5.5 (1494). Higher on the full board are Claude Fable 5 and GPT-5.6 Sol. K3 also tops AA's AutomationBench. For agentic coding among the three, K3 benchmarks best and costs the least.
Does Kimi K3 support image input? Yes, native vision input, text output. GPT-5.5 and Opus 4.8 also take images. Kimi K2.7 Code is text-only, so an image_url call fails on it.
Is Kimi K3 open source? Moonshot announced open weights "by July 27 2026" and calls K3 the first open 3T-class model. As of this writing the weights are not out and the license was not stated, so today you use K3 through a hosted API. Once released it would lead open-weight models on the AA index, ahead of GLM-5.2 and DeepSeek V4 Pro.
How big is Kimi K3? 2.8T total parameters, a Mixture-of-Experts activating 16 of 896 experts per token, with a 1M token context and native vision. Much larger than GLM-5.2 (753B) or DeepSeek V4 Pro (1.6T).
What is the ofox model ID for Kimi K3? moonshotai/kimi-k3 on the OpenAI-compatible endpoint at https://api.ofox.ai/v1. GPT-5.5 is openai/gpt-5.5 and Opus 4.8 is anthropic/claude-opus-4.8, so you A/B all three by swapping one string.
Originally published on ofox.ai/blog.
Top comments (0)