ZMIME
Comparison · 3 models · Updated Nov 18, 2026

Kimi K2.6 vs Qwen3 235B-A22B Thinking 2507 vs Grok 4.20 (Reasoning)

Too close to call on our weighted score (Grok 4.20 (Reasoning) 68, Kimi K2.6 65, Qwen3 235B-A22B Thinking 2507 57). The right pick depends on what you value most.

Price in rating: Included · Exclude price

  1. Moonshot AI

    Kimi K2.6

    Released Apr 20, 2026

    65/100
    • ECI151.6
    • Price$0.95 / $4.00
    • Context262K
  2. Alibaba (Qwen)

    Qwen3 235B-A22B Thinking 2507

    Released Jul 25, 2025

    57/100
    • ECI143.3
    • Price$0.30 / $2.90
    • Context262K
  3. xAI

    Grok 4.20 (Reasoning)

    Released Mar 9, 2026

    68/100
    • ECI152.3
    • Price$1.25 / $2.50
    • Context1M
01 — Verdict

Too close to call

It is close. Our weighted score puts them within 3 points (Grok 4.20 (Reasoning) 68/100, Kimi K2.6 65/100, Qwen3 235B-A22B Thinking 2507 57/100), so choose by what matters most for your work: Grok 4.20 (Reasoning) for raw capability and Qwen3 235B-A22B Thinking 2507 on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

  • CapabilityGrok 4.20 (Reasoning)Capabilities Index (ECI): Grok 4.20 (Reasoning) 152.3 · Kimi K2.6 151.6 · Qwen3 235B-A22B Thinking 2507 143.3
  • Lowest priceQwen3 235B-A22B Thinking 2507Qwen3 235B-A22B Thinking 2507 $0.95 · Grok 4.20 (Reasoning) $1.56 · Kimi K2.6 $1.71 per 1M tokens (3:1 blend)
  • Longest contextGrok 4.20 (Reasoning)Grok 4.20 (Reasoning) 1,000,000 · Kimi K2.6 262,144 · Qwen3 235B-A22B Thinking 2507 262,144 tokens
  • Widest inputsKimi K2.6 and Grok 4.20 (Reasoning)Kimi K2.6: Text, Images, Video · Qwen3 235B-A22B Thinking 2507: Text · Grok 4.20 (Reasoning): Text, Images, PDFs
  • Self-hostingKimi K2.6 and Qwen3 235B-A22B Thinking 2507Publishes downloadable weights (Apache 2.0)
How the score is built
MeasureWeightKimi K2.6Qwen3 235B-A22B Thinking 2507Grok 4.20 (Reasoning)
CapabilityCapabilities Index (ECI)50%797179
Price25%374939
Inputs & features15%803580
Context window10%434364
Overall100%65/10057/10068/100
02 — Side by side

Every spec in one table

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

Kimi K2.6 vs Qwen3 235B-A22B Thinking 2507 vs Grok 4.20 (Reasoning) specifications side by side
SpecificationKimi K2.6Moonshot AIQwen3 235B-A22B Thinking 2507Alibaba (Qwen)Grok 4.20 (Reasoning)xAI
Capability
Capabilities Index (ECI)151.6143.3152.3 (best)
ECI rank#61 of 158#100 of 158#59 of 158 (best)
GPQA DiamondGraduate-level science questions90.8% (best)80.1%89.3%
FrontierMath Tiers 1–3Research-level mathematics57.2% (best)—44.9%
OTIS Mock AIME 2024–2025Competition mathematics96.1% (best)86.7%92.2%
SWE-bench VerifiedFixing real GitHub issues76.7%——
SimpleQA VerifiedShort factual questions34.9%40.4% (best)30.2%
Price per million tokens
Input$0.95$0.30 (best)$1.25
Output$4.00$2.90$2.50 (best)
Cached input$0.16 (best)—$0.20
Blended (3:1)$1.71$0.95 (best)$1.56
Long-context rateSame rateSame rateOver 200K: $2.50 / $5.00
Price sourceOfficial Moonshot AI APIMedian of 11 providersOfficial xAI API
Limits
Context window262,144 tokens262,144 tokens1,000,000 tokens (best)
Max output262,144 tokens (best)81,920 tokens30,000 tokens
Inputs and features
TextYesYesYes
ImagesYesNoYes
PDFsNoNoYes
AudioNoNoNo
VideoYesNoNo
ReasoningYesYesYes
Tool callingYesYesYes
Structured outputYesNoYes
Availability
WeightsOpenOpenApache 2.0Proprietary
API model IDkimi-k2.6—grok-4.20-0309-reasoning
API providers64 (best)1214
ReleasedApr 20, 2026Jul 25, 2025Mar 9, 2026
Knowledge cutoffJan 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.

  • Kimi K2.6$17.50
  • Qwen3 235B-A22B Thinking 2507$8.80
  • Grok 4.20 (Reasoning)$17.50
04 — Questions

Which should you choose?

Which is better: Kimi K2.6, Qwen3 235B-A22B Thinking 2507 or Grok 4.20 (Reasoning)?

It is close. Our weighted score puts them within 3 points (Grok 4.20 (Reasoning) 68/100, Kimi K2.6 65/100, Qwen3 235B-A22B Thinking 2507 57/100), so choose by what matters most for your work: Grok 4.20 (Reasoning) for raw capability and Qwen3 235B-A22B Thinking 2507 on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

Which is cheaper, Kimi K2.6, Qwen3 235B-A22B Thinking 2507 or Grok 4.20 (Reasoning)?

Qwen3 235B-A22B Thinking 2507 is cheaper at $0.30 input / $2.90 output per million tokens (median across 11 API providers). Grok 4.20 (Reasoning) costs $1.25 input / $2.50 output per million tokens (official xAI API price); Kimi K2.6 costs $0.95 input / $4.00 output per million tokens (official Moonshot AI API price). At a typical mix of three input tokens to one output token, that is $0.95 per million tokens for Qwen3 235B-A22B Thinking 2507 versus $1.56 for Grok 4.20 (Reasoning) (1.6× as much) and $1.71 for Kimi K2.6 (1.8× as much).

Which scores higher on benchmarks?

Grok 4.20 (Reasoning) scores higher on the Capabilities Index (ECI): Grok 4.20 (Reasoning) 152.3 (#59 of 158), Kimi K2.6 151.6 (#61 of 158) and Qwen3 235B-A22B Thinking 2507 143.3 (#100 of 158). The confidence ranges of the top two overlap (150.6–154.1 vs 149.9–153.4), so treat the gap as small. On individual benchmarks: GPQA Diamond — Kimi K2.6 90.8%, Grok 4.20 (Reasoning) 89.3%, Qwen3 235B-A22B Thinking 2507 80.1%; OTIS Mock AIME 2024–2025 — Kimi K2.6 96.1%, Grok 4.20 (Reasoning) 92.2%, Qwen3 235B-A22B Thinking 2507 86.7%; SimpleQA Verified — Qwen3 235B-A22B Thinking 2507 40.4%, Kimi K2.6 34.9%, Grok 4.20 (Reasoning) 30.2%.

Which is better for coding?

There are no published SWE-bench Verified results for Qwen3 235B-A22B Thinking 2507 and Grok 4.20 (Reasoning) yet, so there is no like-for-like coding score. On overall capability, Grok 4.20 (Reasoning) 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?

Grok 4.20 (Reasoning) has the largest context window at 1,000,000 tokens, against 262,144 for Kimi K2.6 and 262,144 for Qwen3 235B-A22B Thinking 2507. Maximum output per response: Kimi K2.6 up to 262,144, Qwen3 235B-A22B Thinking 2507 up to 81,920, Grok 4.20 (Reasoning) up to 30,000 tokens.

Which can read images, PDFs, audio or video?

Kimi K2.6 accepts text, images and video; Qwen3 235B-A22B Thinking 2507 accepts text; Grok 4.20 (Reasoning) accepts text, images and PDFs. Kimi K2.6 handles the widest range of inputs.

Are any of these open source?

Kimi K2.6 and Qwen3 235B-A22B Thinking 2507 publishes its weights (Apache 2.0) and can be self-hosted; Grok 4.20 (Reasoning) is proprietary.

Which is newer?

Kimi K2.6 is the newest, released Apr 20, 2026. Grok 4.20 (Reasoning) came out Mar 9, 2026; Qwen3 235B-A22B Thinking 2507 came out Jul 25, 2025. Knowledge cutoff: Kimi K2.6 Jan 2025.

How do you decide the winner?

This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, 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.