ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Claude Fable 5 vs GPT-5.4 vs Kimi K3

Claude Fable 5 comes out ahead, 85 to 82 and 81 on our weighted score, though GPT-5.4 is 3.6× cheaper per token.

Price in rating: Excluded · Include price

  1. Our pick

    Anthropic

    Claude Fable 5

    Released Jun 9, 2026

    85/100
    • ECI164.2
    • Price$10.00 / $50.00
    • Context1M
  2. OpenAI

    GPT-5.4

    Released Mar 5, 2026

    81/100
    • ECI157.9
    • Price$2.50 / $15.00
    • Context1.05M
  3. Moonshot AI

    Kimi K3

    Released Jul 16, 2026

    82/100
    • ECI158.6
    • Price$3.00 / $15.00
    • Context1.05M
01 — Verdict

Claude Fable 5 is our pick

Claude Fable 5 is the better all-round choice, scoring 85/100 against Kimi K3 (82) and GPT-5.4 (81). It leads on capability. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityClaude Fable 5Capabilities Index (ECI): Claude Fable 5 164.2 · Kimi K3 158.6 · GPT-5.4 157.9
  • Lowest priceGPT-5.4GPT-5.4 $5.63 · Kimi K3 $6.00 · Claude Fable 5 $20.00 per 1M tokens (3:1 blend)
  • Longest contextGPT-5.4 and Kimi K3GPT-5.4 1,050,000 · Kimi K3 1,048,576 · Claude Fable 5 1,000,000 tokens
  • Widest inputsSame inputsClaude Fable 5: Text, Images, PDFs · GPT-5.4: Text, Images, PDFs · Kimi K3: Text, Images, Video
  • Self-hostingKimi K3Publishes downloadable weights
How the score is built
MeasureWeightClaude Fable 5GPT-5.4Kimi K3
CapabilityCapabilities Index (ECI)67%918585
Inputs & features20%808080
Context window13%646565
Overall100%85/10081/10082/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Dashes mean the data is not published.

Claude Fable 5 vs GPT-5.4 vs Kimi K3 specifications side by side
SpecificationClaude Fable 5AnthropicGPT-5.4OpenAIKimi K3Moonshot AI
Capability
Capabilities Index (ECI)164.2 (best)157.9158.6
ECI rank#12 of 158 (best)#28 of 158#25 of 158
GPQA DiamondGraduate-level science questions93.9% (best)93.3%93.1%
FrontierMath Tiers 1–3Research-level mathematics87.0% (best)78.6%72.2%
OTIS Mock AIME 2024–2025Competition mathematics100% (best)97.8%97.2%
SWE-bench VerifiedFixing real GitHub issues—76.9%—
SimpleQA VerifiedShort factual questions70.7% (best)45.1%50.6%
Price per million tokens
Input$10.00$2.50 (best)$3.00
Output$50.00$15.00 (best)$15.00 (best)
Cached input$1.00$0.25 (best)$0.30
Blended (3:1)$20.00$5.63 (best)$6.00
Long-context rateSame rateOver 272K: $5.00 / $22.50Same rate
Price sourceOfficial Anthropic APIOfficial OpenAI APIOfficial Moonshot AI API
Limits
Context window1,000,000 tokens1,050,000 tokens (best)1,048,576 tokens
Max output128,000 tokens128,000 tokens1,048,576 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesYesYes
PDFsYesYesNo
AudioNoNoNo
VideoNoNoYes
ReasoningYeslow · medium · high · xhigh · maxYeslow · medium · high · xhighYeslow · high · max
Tool callingYesYesYes
Structured outputYesYesYes
Availability
WeightsProprietaryProprietaryOpen
API model IDclaude-fable-5gpt-5.4kimi-k3
API providers555094 (best)
ReleasedJun 9, 2026Mar 5, 2026Jul 16, 2026
Knowledge cutoffJan 31, 2026Aug 31, 2025—
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • Claude Fable 5$200.00
  • GPT-5.4$55.00
  • Kimi K3$60.00
04 — Questions

Which should you choose?

Which is better: Claude Fable 5, GPT-5.4 or Kimi K3?

Claude Fable 5 is the better all-round choice, scoring 85/100 against Kimi K3 (82) and GPT-5.4 (81). It leads on capability. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, Claude Fable 5, GPT-5.4 or Kimi K3?

GPT-5.4 is cheaper at $2.50 input / $15.00 output per million tokens (official OpenAI API price). Kimi K3 costs $3.00 input / $15.00 output per million tokens (official Moonshot AI API price); Claude Fable 5 costs $10.00 input / $50.00 output per million tokens (official Anthropic API price). At a typical mix of three input tokens to one output token, that is $5.63 per million tokens for GPT-5.4 versus $6.00 for Kimi K3 (1.1× as much) and $20.00 for Claude Fable 5 (3.6× as much).

Which scores higher on benchmarks?

Claude Fable 5 scores higher on the Capabilities Index (ECI): Claude Fable 5 164.2 (#12 of 158), Kimi K3 158.6 (#25 of 158) and GPT-5.4 157.9 (#28 of 158). Their confidence ranges do not overlap (161.8–166.9 vs 156.6–160.8), so the gap is a real one. On individual benchmarks: GPQA Diamond — Claude Fable 5 93.9%, GPT-5.4 93.3%, Kimi K3 93.1%; FrontierMath Tiers 1–3 — Claude Fable 5 87.0%, GPT-5.4 78.6%, Kimi K3 72.2%; OTIS Mock AIME 2024–2025 — Claude Fable 5 100%, GPT-5.4 97.8%, Kimi K3 97.2%; SimpleQA Verified — Claude Fable 5 70.7%, Kimi K3 50.6%, GPT-5.4 45.1%.

Which is better for coding?

There are no published SWE-bench Verified results for Claude Fable 5 and Kimi K3 yet, so there is no like-for-like coding score. On overall capability, Claude Fable 5 leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

GPT-5.4 and Kimi K3 have the largest context windows (1,050,000 and 1,048,576 tokens), against 1,000,000 for Claude Fable 5. Maximum output per response: Claude Fable 5 up to 128,000, GPT-5.4 up to 128,000, Kimi K3 up to 1,048,576 tokens.

Which can read images, PDFs, audio or video?

Claude Fable 5 accepts text, images and PDFs; GPT-5.4 accepts text, images and PDFs; Kimi K3 accepts text, images and video. They handle the same number of input types.

Are any of these open source?

Kimi K3 publishes its weights and can be self-hosted; Claude Fable 5 and GPT-5.4 is proprietary.

Which is newer?

Kimi K3 is the newest, released Jul 16, 2026. Claude Fable 5 came out Jun 9, 2026; GPT-5.4 came out Mar 5, 2026. Knowledge cutoff: Claude Fable 5 Jan 31, 2026, GPT-5.4 Aug 31, 2025.

How do you decide the winner?

This rating uses capability (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 18, 2026.