Skip to main content
BenchLM
Advertise

Model comparison

Grok 4.3 vs Qwen3.5 397B

Updated July 30, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Grok 4.3

xAI

64.2/100

Supported · Public rank #36

90% interval 54.5–74.0

Qwen3.5 397B

Alibaba

56.2/100

Estimated · Public rank #78

90% interval 44.7–67.7

Grok 4.3 has the higher public score estimate, 64.2 versus 56.19, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

2 results are shared. Category rows based on different benchmark sets are marked directional and do not name a winner.

Which one for your work

Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.

  • Long documents

    Prompts that approach the documented context limit

    Grok 4.3

    Grok 4.3 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Qwen3.5 397B

    Qwen3.5 397B has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    Qwen3.5 397B

    Qwen3.5 397B has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    No shared weighted benchmark basis supports a winner.

    Confidence: limited

  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Not enough matched evidence

    The page does not recommend a cost winner because at least one model cannot fit the stated workload in one request. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

Shared results
2
Grok 4.3 only
0
Qwen3.5 397B only
36
Like-for-like categories
0 / 8

Category results, on a stated basis

Each row states whether both averages use the same weighted benchmark set. Directional and not-comparable rows remain visible, but they never receive a winner in this template.

Agentic

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
56.5
Weighted basis
0 vs 2 rows
Reading
Not comparable

Coding

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
66.5
Weighted basis
0 vs 2 rows
Reading
Not comparable

Reasoning

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
63.2
Weighted basis
0 vs 1 rows
Reading
Not comparable

Knowledge

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
56.6
Weighted basis
0 vs 4 rows
Reading
Not comparable

Math

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
90.6
Weighted basis
0 vs 2 rows
Reading
Not comparable

Multilingual

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
84.7
Weighted basis
0 vs 1 rows
Reading
Not comparable

Multimodal

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
79.6
Weighted basis
0 vs 2 rows
Reading
Not comparable

Instruction following

Not comparable
Grok 4.3
Not measured
Qwen3.5 397B
92.6
Weighted basis
0 vs 1 rows
Reading
Not comparable

Shape of the matched evidence

Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.

A shared-evidence shape is not available.

BenchLM does not draw a radar or infer missing axes when the matched evidence is too sparse.

What each workload costs

Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.

Chat turn

1K fresh input + 500 output tokens

Grok 4.3
$0.0025
Fits in one request
Qwen3.5 397B
$0.0024
Fits in one request

Qwen3.5 397B has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Grok 4.3
$0.07
Fits in one request
Qwen3.5 397B
$0.0408
Fits in one request

Qwen3.5 397B has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

Grok 4.3
$0.09
Fits in one request
Qwen3.5 397B
$0.168
Does not fit in one request
Cached input priced at the published list-input rate

Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate.

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Context window

Maximum documented context; output-token limits may be lower.

Grok 4.3

1M

Qwen3.5 397B

128K

API model ID

Grok 4.3

Not sourced

Qwen3.5 397B

Not sourced

Cached-input rate

A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.

Grok 4.3

$0.2 per 1M cached input tokens

Qwen3.5 397B

Not published

Documented inputs

Grok 4.3

Not sourced

Qwen3.5 397B

Not sourced

Documented outputs

Grok 4.3

Not sourced

Qwen3.5 397B

Not sourced

Provider availability

Grok 4.3

Not sourced

Qwen3.5 397B

Not sourced

Reasoning profile

Grok 4.3

Reasoning

Qwen3.5 397B

Non-Reasoning

Weight access

Grok 4.3

Proprietary

Qwen3.5 397B

Open Weight

License

Grok 4.3

Proprietary

Qwen3.5 397B

Open Weight

Release date

Grok 4.3

2026-04-30

Qwen3.5 397B

2026-02-16

If you already use one of these models
Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
Quality signal
Grok 4.3 has the higher public score estimate, 64.2 versus 56.19, but the 90% score intervals overlap.
Workload cost
Repository review: $0.07 vs $0.0408. Cache-heavy agent loop: $0.09 vs $0.168.
Context tradeoff
Grok 4.3 has the larger documented window (1M).

Run the same representative tasks against both endpoints before changing production traffic.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence38 rows

Agentic

  • Gert Labs

    Shared source
    Grok 4.343.86%
    Qwen3.5 397B46.76%

    Qwen3.5 397B leads this result

  • ResearchClawBench

    Shared source
    Grok 4.312.4%
    Qwen3.5 397B14.2%

    Qwen3.5 397B leads this result

  • Terminal-Bench 2.0

    Grok 4.3—
    Qwen3.5 397B52.5%
    Source

    Not directly comparable

  • BrowseComp

    Grok 4.3—
    Qwen3.5 397B62%
    Source

    Not directly comparable

  • Claw-Eval

    Grok 4.3—
    Qwen3.5 397B56.8%
    Source

    Not directly comparable

  • QwenClawBench

    Grok 4.3—
    Qwen3.5 397B51.8%
    Source

    Not directly comparable

  • τ³-bench results

    Grok 4.3—
    Qwen3.5 397B68.4%
    Source

    Not directly comparable

  • VITA-Bench

    Grok 4.3—
    Qwen3.5 397B43.7%
    Source

    Not directly comparable

  • DeepPlanning

    Grok 4.3—
    Qwen3.5 397B37.6%
    Source

    Not directly comparable

  • Toolathlon

    Grok 4.3—
    Qwen3.5 397B36.3%
    Source

    Not directly comparable

  • MCP Atlas

    Grok 4.3—
    Qwen3.5 397B46.1%
    Source

    Not directly comparable

  • MCP-Tasks

    Grok 4.3—
    Qwen3.5 397B74.2%
    Source

    Not directly comparable

  • WideResearch

    Grok 4.3—
    Qwen3.5 397B74.0%
    Source

    Not directly comparable

Coding

  • SWE-bench Verified

    Grok 4.3—
    Qwen3.5 397B76.2%
    Source

    Not directly comparable

  • LiveCodeBench v6

    Grok 4.3—
    Qwen3.5 397B83.6%
    Source

    Not directly comparable

  • SWE-bench Pro

    Grok 4.3—
    Qwen3.5 397B50.9%
    Source

    Not directly comparable

Reasoning

  • LongBench v2

    Grok 4.3—
    Qwen3.5 397B63.2%
    Source

    Not directly comparable

  • AI-Needle

    Grok 4.3—
    Qwen3.5 397B68.7%
    Source

    Not directly comparable

Knowledge

  • GPQA

    Grok 4.3—
    Qwen3.5 397B88.4%
    Source

    Not directly comparable

  • SuperGPQA

    Grok 4.3—
    Qwen3.5 397B70.4%
    Source

    Not directly comparable

  • MMLU-Pro

    Grok 4.3—
    Qwen3.5 397B87.8%
    Source

    Not directly comparable

  • MMLU-Redux

    Grok 4.3—
    Qwen3.5 397B94.9%
    Source

    Not directly comparable

  • C-Eval

    Grok 4.3—
    Qwen3.5 397B93%
    Source

    Not directly comparable

  • HLE

    Grok 4.3—
    Qwen3.5 397B28.7%
    Source

    Not directly comparable

Math

  • AIME26

    Grok 4.3—
    Qwen3.5 397B93.3%
    Source

    Not directly comparable

  • HMMT Feb 2025

    Grok 4.3—
    Qwen3.5 397B94.8%
    Source

    Not directly comparable

  • HMMT Nov 2025

    Grok 4.3—
    Qwen3.5 397B92.7%
    Source

    Not directly comparable

  • HMMT Feb 2026

    Grok 4.3—
    Qwen3.5 397B87.9%
    Source

    Not directly comparable

  • MMAnswerBench

    Grok 4.3—
    Qwen3.5 397B80.9%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    Grok 4.3—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • NOVA-63

    Grok 4.3—
    Qwen3.5 397B59.1%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    Grok 4.3—
    Qwen3.5 397B79%
    Source

    Not directly comparable

  • MathVision

    Grok 4.3—
    Qwen3.5 397B88.6%
    Source

    Not directly comparable

  • CharXiv

    Grok 4.3—
    Qwen3.5 397B80.8%
    Source

    Not directly comparable

  • VideoMMMU

    Grok 4.3—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • ScreenSpot Pro

    Grok 4.3—
    Qwen3.5 397B65.6%
    Source

    Not directly comparable

  • V*

    Grok 4.3—
    Qwen3.5 397B95.8%
    Source

    Not directly comparable

Instruction following

  • IFEval

    Grok 4.3—
    Qwen3.5 397B92.6%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Grok 4.3 or Qwen3.5 397B?

Grok 4.3 has the higher public score estimate, 64.2 versus 56.19, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, Grok 4.3 or Qwen3.5 397B?

The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.

Which is better for agentic tasks, Grok 4.3 or Qwen3.5 397B?

The published evidence does not provide a shared weighted agentic tasks basis for both models, so BenchLM does not name a agentic tasks winner.

Which costs less, Grok 4.3 or Qwen3.5 397B?

For the stated presets, chat costs $0.0025 on Grok 4.3 and $0.0024 on Qwen3.5 397B; repository review costs $0.07 and $0.0408; the cache-heavy agent loop costs $0.09 and $0.168. Qwen3.5 397B does not fit this workload in one request. Qwen3.5 397B has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, Grok 4.3 or Qwen3.5 397B?

Grok 4.3 has the larger documented context window: 1M, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated July 30, 2026

Watch Grok 4.3 vs Qwen3.5 397B

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 2,000+ readers.