Skip to main content
BenchLM
Advertise

Model comparison

DeepSeek V3.2 vs MiniMax M2.7

Updated August 1, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

20 confirmed releases in the last 30 daysSee provider release alerts
DeepSeek V3.2

DeepSeek

54.5/100

Supported · Public rank #89

90% interval 37.7–71.2

MiniMax M2.7

MiniMax

63.1/100

Supported · Public rank #40

90% interval 56.5–69.7

MiniMax M2.7 has the higher public score estimate, 63.09 versus 54.45, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.

4 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

    MiniMax M2.7

    MiniMax M2.7 has the larger documented context window.

    Confidence: documented

  • Chat turn cost

    1K fresh input + 500 output tokens

    DeepSeek V3.2

    DeepSeek V3.2 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

    DeepSeek V3.2

    DeepSeek V3.2 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

    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. DeepSeek V3.2 does not fit this workload in one request. MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 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
4
DeepSeek V3.2 only
3
MiniMax M2.7 only
14
Like-for-like categories
0 / 8

1 category uses 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.

Coding

Directional only
DeepSeek V3.2
60.9
MiniMax M2.7
53.3
Weighted basis
1 vs 2 rows
Reading
Directional only

Agentic

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
57.0
Weighted basis
0 vs 1 rows
Reading
Not comparable

Reasoning

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Knowledge

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Math

Not comparable
DeepSeek V3.2
17.1
MiniMax M2.7
Not measured
Weighted basis
2 vs 0 rows
Reading
Not comparable

Multilingual

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Multimodal

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
Not measured
Weighted basis
0 vs 0 rows
Reading
Not comparable

Instruction following

Not comparable
DeepSeek V3.2
Not measured
MiniMax M2.7
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.

  • SWE-Rebench

    Coding

    DeepSeek V3.2: 60.9%MiniMax M2.7: 51.9%Normalized gap 9.0Shared source

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

DeepSeek V3.2
$0.00049
Fits in one request
MiniMax M2.7
$0.0009
Fits in one request

DeepSeek V3.2 has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

DeepSeek V3.2
$0.01526
Fits in one request
MiniMax M2.7
$0.0186
Fits in one request

DeepSeek V3.2 has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

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

DeepSeek V3.2
$0.0154
Does not fit in one request
MiniMax M2.7
$0.078
Does not fit in one request
Cached input priced at the published list-input rate

DeepSeek V3.2 does not fit this workload in one request. MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 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.

DeepSeek V3.2

128K

MiniMax M2.7

200K

API model ID

DeepSeek V3.2

Not sourced

MiniMax M2.7

Not sourced

Cached-input rate

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

DeepSeek V3.2

$0.028 per 1M cached input tokens

MiniMax M2.7

Not published

Documented inputs

DeepSeek V3.2

Not sourced

MiniMax M2.7

Not sourced

Documented outputs

DeepSeek V3.2

Not sourced

MiniMax M2.7

Not sourced

Provider availability

DeepSeek V3.2

Not sourced

MiniMax M2.7

Not sourced

Reasoning profile

DeepSeek V3.2

Non-Reasoning

MiniMax M2.7

Non-Reasoning

Weight access

DeepSeek V3.2

Open Weight

MiniMax M2.7

Open Weight

License

DeepSeek V3.2

Open Weight

MiniMax M2.7

Open Weight

Release date

DeepSeek V3.2

2025-12-01

MiniMax M2.7

2026-03-18

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
MiniMax M2.7 has the higher public score estimate, 63.09 versus 54.45, but the 90% score intervals overlap.
Workload cost
Repository review: $0.01526 vs $0.0186. Cache-heavy agent loop: $0.0154 vs $0.078.
Context tradeoff
MiniMax M2.7 has the larger documented window (200K).

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 evidence21 rows

Agentic

  • Claw-Eval

    Shared source
    DeepSeek V3.240.2%
    MiniMax M2.748.7%

    MiniMax M2.7 leads this result

  • VITA-Bench

    DeepSeek V3.218.5%
    Source
    MiniMax M2.7—

    Not directly comparable

  • Gert Labs

    Shared source
    DeepSeek V3.229.57%
    MiniMax M2.740.40%

    MiniMax M2.7 leads this result

  • Terminal-Bench 2.0

    DeepSeek V3.2—
    MiniMax M2.757%
    Source

    Not directly comparable

  • Toolathlon

    DeepSeek V3.2—
    MiniMax M2.746.3%
    Source

    Not directly comparable

  • MLE-Bench Lite

    DeepSeek V3.2—
    MiniMax M2.766.6%
    Source

    Not directly comparable

  • MM-ClawBench

    DeepSeek V3.2—
    MiniMax M2.762.7%
    Source

    Not directly comparable

Coding

  • SWE-Rebench

    Shared source
    DeepSeek V3.260.9%
    MiniMax M2.751.9%

    DeepSeek V3.2 leads this result

  • React Native Evals

    Shared source
    DeepSeek V3.271.5%
    MiniMax M2.771.4%

    DeepSeek V3.2 leads this result

  • SWE-bench Verified*

    DeepSeek V3.2—
    MiniMax M2.775.4%
    Source

    Not directly comparable

  • SWE-bench Pro

    DeepSeek V3.2—
    MiniMax M2.756.2%
    Source

    Not directly comparable

  • SWE Multilingual

    DeepSeek V3.2—
    MiniMax M2.776.5%
    Source

    Not directly comparable

  • Multi-SWE Bench

    DeepSeek V3.2—
    MiniMax M2.752.7%
    Source

    Not directly comparable

  • VIBE-Pro

    DeepSeek V3.2—
    MiniMax M2.755.6%
    Source

    Not directly comparable

  • NL2Repo

    DeepSeek V3.2—
    MiniMax M2.739.8%
    Source

    Not directly comparable

  • Vibe Code Bench

    DeepSeek V3.2—
    MiniMax M2.727.04%
    Source

    Not directly comparable

Knowledge

  • GPQA-D

    DeepSeek V3.2—
    MiniMax M2.787.0%
    Source

    Not directly comparable

  • MMLU-Pro (Arcee)

    DeepSeek V3.2—
    MiniMax M2.780.8%
    Source

    Not directly comparable

Math

  • FrontierMath v2 (Tiers 1-3)

    DeepSeek V3.222.100%
    Source
    MiniMax M2.7—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    DeepSeek V3.22.100%
    Source
    MiniMax M2.7—

    Not directly comparable

  • AIME25 (Arcee)

    DeepSeek V3.2—
    MiniMax M2.780.0%
    Source

    Not directly comparable

Frequently asked questions

Which is better, DeepSeek V3.2 or MiniMax M2.7?

MiniMax M2.7 has the higher public score estimate, 63.09 versus 54.45, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.

Which is better for coding, DeepSeek V3.2 or MiniMax M2.7?

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, DeepSeek V3.2 or MiniMax M2.7?

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, DeepSeek V3.2 or MiniMax M2.7?

For the stated presets, chat costs $0.00049 on DeepSeek V3.2 and $0.0009 on MiniMax M2.7; repository review costs $0.01526 and $0.0186; the cache-heavy agent loop costs $0.0154 and $0.078. DeepSeek V3.2 does not fit this workload in one request. MiniMax M2.7 does not fit this workload in one request. MiniMax M2.7 has no published cached-input rate, so cached tokens use its listed input rate.

Which has the larger context window, DeepSeek V3.2 or MiniMax M2.7?

MiniMax M2.7 has the larger documented context window: 200K, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated August 1, 2026

Watch DeepSeek V3.2 vs MiniMax M2.7

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

Read a sample issue

Join 2,000+ readers.