Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start the free Radar Brief
BenchLM
xAI logo
Model A
Grok 4.20

xAI

55.44/100

Estimated · Public rank #100

90% interval 38.5–72.3

Grok 4.20 vs Step 3.7 Flash

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

StepFun logo
Model B
Step 3.7 Flash

StepFun

51.05/100

Estimated · Public rank #126

90% interval 39.5–62.6

Decision reading

Grok 4.20 has the higher public score estimate, 55.44 versus 51.05, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

5 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 4.20

    Grok 4.20 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    Step 3.7 Flash

    Step 3.7 Flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates

  • Cache-heavy agent loop cost

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

    Step 3.7 Flash

    Step 3.7 Flash has the lower estimated token cost for this stated workload. Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate. Step 3.7 Flash has no published cached-input rate, so cached tokens use its listed input rate.

    Confidence: rate-fallback

Show secondary and unsupported calls

  • Repository review cost

    50K fresh input + 3K output tokens

    Step 3.7 Flash

    Step 3.7 Flash 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

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    The category averages use different weighted benchmark sets, so they are directional rather than like-for-like.

    Confidence: limited

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
5
Grok 4.20 only
13
Step 3.7 Flash only
6
Like-for-like categories
0 / 8

2 categories use different evidence sets. Those rows remain visible for coverage context but do not name a winner.

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

Directional only
Grok 4.20
47.1
Step 3.7 Flash
66.4
Weighted basis
1 vs 2 rows
Reading
Directional only

Coding

Directional only
Grok 4.20
67.1
Step 3.7 Flash
56.3
Weighted basis
2 vs 1 rows
Reading
Directional only

Reasoning

Not comparable
Grok 4.20
53.3
Step 3.7 Flash
Not measured
Weighted basis
1 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
Grok 4.20
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Math

Not comparable
Grok 4.20
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
Grok 4.20
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
Grok 4.20
70.1
Step 3.7 Flash
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
Grok 4.20
Not measured
Step 3.7 Flash
Not measured
Weighted basis
0 vs 0 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.

Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.

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 4.20
$0.005
Fits in one request
Step 3.7 Flash
$0.00077
Fits in one request

Step 3.7 Flash has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

Grok 4.20
$0.118
Fits in one request
Step 3.7 Flash
$0.01345
Fits in one request

Step 3.7 Flash 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 4.20
$0.5
Fits in one request
Cached input priced at the published list-input rate
Step 3.7 Flash
$0.0555
Fits in one request
Cached input priced at the published list-input rate

Step 3.7 Flash has the lower modeled cost

Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate. Step 3.7 Flash 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 4.20

2M

Step 3.7 Flash

256K

API model ID

Grok 4.20

Not sourced

Step 3.7 Flash

Not sourced

Cached-input rate

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

Grok 4.20

Not published

Step 3.7 Flash

Not published

Documented inputs

Grok 4.20

Not sourced

Step 3.7 Flash

Not sourced

Documented outputs

Grok 4.20

Not sourced

Step 3.7 Flash

Not sourced

Provider availability

Grok 4.20

Not sourced

Step 3.7 Flash

Not sourced

Reasoning profile

Grok 4.20

Reasoning

Step 3.7 Flash

Reasoning

Weight access

Grok 4.20

Proprietary

Step 3.7 Flash

Open Weight

License

Grok 4.20

Proprietary

Step 3.7 Flash

Open Weight

Release date

Grok 4.20

2026-03-10

Step 3.7 Flash

2026-05-29

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 4.20 has the higher public score estimate, 55.44 versus 51.05, but the 90% score intervals overlap.
Workload cost
Repository review: $0.118 vs $0.01345. Cache-heavy agent loop: $0.5 vs $0.0555.
Context tradeoff
Grok 4.20 has the larger documented window (2M).

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 evidence24 rows

Agentic

  • Terminal-Bench 2.0

    Grok 4.2047.1%
    Source
    Step 3.7 Flash59.5%
    Source

    Step 3.7 Flash leads this result

  • DeepSearchQA

    Grok 4.2062.8%
    Source
    Step 3.7 Flash92.8%
    Source

    Step 3.7 Flash leads this result

  • Gert Labs

    Shared source
    Grok 4.2038.36%
    Step 3.7 Flash51.57%

    Step 3.7 Flash leads this result

  • BrowseComp

    Grok 4.20—
    Step 3.7 Flash75.8%
    Source

    Not directly comparable

  • Toolathlon

    Grok 4.20—
    Step 3.7 Flash49.5%
    Source

    Not directly comparable

  • Claw-Eval

    Grok 4.20—
    Step 3.7 Flash67.1%
    Source

    Not directly comparable

  • HLE w/ tools

    Grok 4.20—
    Step 3.7 Flash47.2%
    Source

    Not directly comparable

Coding

  • LiveCodeBench Pro

    Grok 4.2074.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SWE-bench Verified

    Grok 4.2076.7%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SWE-bench Pro

    Grok 4.2051.8%
    Source
    Step 3.7 Flash56.3%
    Source

    Step 3.7 Flash leads this result

  • Vibe Code Bench

    Grok 4.204.06%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • Terminal-Bench 2.0

    Grok 4.20—
    Step 3.7 Flash59.5%
    Source

    Not directly comparable

Reasoning

  • ARC-AGI-2

    Grok 4.2053.3%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • ARC-AGI-3

    Grok 4.200.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

Knowledge

  • GPQA-D

    Grok 4.2088.5%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • HLE w/o tools

    Grok 4.2031.6%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • HealthBench Hard

    Grok 4.2020.3%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • MedXpertQA (Text)

    Grok 4.2050.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

Multimodal

  • MMMU-Pro

    Grok 4.2075.2%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • CharXiv

    Grok 4.2060.9%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • ERQA

    Grok 4.2054.1%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • SimpleVQA

    Grok 4.2057.4%
    Source
    Step 3.7 Flash79.2%
    Source

    Step 3.7 Flash leads this result

  • MedXpertQA (MM)

    Grok 4.2065.8%
    Source
    Step 3.7 Flash—

    Not directly comparable

  • V*

    Grok 4.20—
    Step 3.7 Flash95.3%
    Source

    Not directly comparable

Frequently asked questions

Which is better, Grok 4.20 or Step 3.7 Flash?

Grok 4.20 has the higher public score estimate, 55.44 versus 51.05, 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 4.20 or Step 3.7 Flash?

The current coding averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, Grok 4.20 or Step 3.7 Flash?

The current agentic tasks averages use different weighted benchmark sets, so BenchLM does not name a winner from them. Read the shared benchmark rows directly and test the models on the same task set.

Which costs less, Grok 4.20 or Step 3.7 Flash?

For the stated presets, chat costs $0.005 on Grok 4.20 and $0.00077 on Step 3.7 Flash; repository review costs $0.118 and $0.01345; the cache-heavy agent loop costs $0.5 and $0.0555. Grok 4.20 has no published cached-input rate, so cached tokens use its listed input rate. Step 3.7 Flash has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, Grok 4.20 or Step 3.7 Flash?

Grok 4.20 has the larger documented context window: 2M, compared with 256K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 30, 2026

Watch Grok 4.20 vs Step 3.7 Flash

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

Read a sample issue

Join 2,000+ readers.