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.

See the free Radar Brief
BenchLM
OpenAI logo
Model A
GPT-5.5

OpenAI

73.27/100

Supported · Public rank #9

90% interval 71.0–75.6

GPT-5.5 vs Qwen3.5 397B

Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.

Alibaba logo
Model B
Qwen3.5 397B

Alibaba

56.44/100

Estimated · Public rank #97

90% interval 44.9–68.0

Decision reading

GPT-5.5 has the higher public score, 73.27 versus 56.44, and the 90% score intervals do not overlap.

10 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Share or export

Share on XLinkedInSocial cardCSVJSON

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

    GPT-5.5

    GPT-5.5 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

    Qwen3.5 397B is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like.

    Confidence: limited

  • Agentic work

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

    Not enough matched evidence

    Qwen3.5 397B is scored on Estimated evidence for agentic, so the reading is directional rather than like-for-like.

    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
10
GPT-5.5 only
28
Qwen3.5 397B only
28
Like-for-like categories
0 / 8

5 categories rest on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.

Category results, on a stated basis

Each row shows the public-lane category score for both models: the BenchAlign lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.

Agentic

Directional only
GPT-5.5
63.9
Supported · #15/151
Qwen3.5 397B
48.2
Estimated · #77/151
Basis
BenchAlign lane · 13 vs 13 public rows
Reading
Directional only

Coding

Directional only
GPT-5.5
67.7
Supported · #8/183
Qwen3.5 397B
52.4
Estimated · #56/183
Basis
BenchAlign lane · 9 vs 3 public rows
Reading
Directional only

Knowledge

Directional only
GPT-5.5
73.3
Supported · #7/181
Qwen3.5 397B
51.6
Estimated · #80/181
Basis
BenchAlign lane · 6 vs 6 public rows
Reading
Directional only

Multimodal

Directional only
GPT-5.5
71.3
#19/48
Qwen3.5 397B
63.0
#27/48
Basis
Provisional lane · 2 vs 2 weighted rows
Reading
Directional only

Instruction following

Directional only
GPT-5.5
92.9
#7/120
Qwen3.5 397B
0.0
#120/120
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Directional only

Reasoning

Not comparable
GPT-5.5
63.5
#15/22
Qwen3.5 397B
59.8
Unranked · 2 rankable rows
Basis
Provisional lane · 2 vs 1 weighted rows
Reading
Not comparable

Math

Not comparable
GPT-5.5
69.6
Unranked · 3 rankable rows
Qwen3.5 397B
74.2
Unranked · 5 rankable rows
Basis
Provisional lane · 2 vs 2 weighted rows
Reading
Not comparable

Multilingual

Not comparable
GPT-5.5
Not ranked
Qwen3.5 397B
69.7
#5/12
Basis
Provisional lane · 0 vs 1 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign) differ from the provisional-lane categories. Unranked scores sit on the lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.

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

GPT-5.5
$0.02
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

GPT-5.5
$0.34
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

GPT-5.5
$0.5
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.

Qwen3.5 397B

128K

Cached-input rate

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

GPT-5.5

$0.5 per 1M cached input tokens

OpenAI pricing

Qwen3.5 397B

Not published

Documented inputs

GPT-5.5

Not sourced

Qwen3.5 397B

Not sourced

Documented outputs

GPT-5.5

Not sourced

Qwen3.5 397B

Not sourced

Provider availability

GPT-5.5

Not sourced

Qwen3.5 397B

Not sourced

Reasoning profile

GPT-5.5

Reasoning

Qwen3.5 397B

Non-Reasoning

Weight access

GPT-5.5

Proprietary

Qwen3.5 397B

Open Weight

License

GPT-5.5

Proprietary

Qwen3.5 397B

Open Weight

Release date

GPT-5.5

2026-04-23

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
GPT-5.5 has the higher public score, 73.27 versus 56.44, and the 90% score intervals do not overlap.
Workload cost
Repository review: $0.34 vs $0.0408. Cache-heavy agent loop: $0.5 vs $0.168.
Context tradeoff
GPT-5.5 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 evidence66 rows

Agentic

  • Terminal-Bench 2.0

    GPT-5.582%
    Source
    Qwen3.5 397B52.5%
    Source

    GPT-5.5 leads this result

  • CyberGym

    GPT-5.581.8%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • BrowseComp

    GPT-5.584.4%
    Source
    Qwen3.5 397B62%
    Source

    GPT-5.5 leads this result

  • OSWorld-Verified

    GPT-5.578.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • MCP Atlas

    GPT-5.575.3%
    Source
    Qwen3.5 397B46.1%
    Source

    GPT-5.5 leads this result

  • Toolathlon

    GPT-5.555.6%
    Source
    Qwen3.5 397B36.3%
    Source

    GPT-5.5 leads this result

  • τ²-bench results

    GPT-5.598%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • Gert Labs

    Shared source
    GPT-5.572.93%
    Qwen3.5 397B46.76%

    GPT-5.5 leads this result

  • ResearchClawBench

    Shared source
    GPT-5.517.0%
    Qwen3.5 397B14.2%

    GPT-5.5 leads this result

  • OSWorld 2.0

    GPT-5.513.0%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • JobBench

    GPT-5.542.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • ExploitGym

    GPT-5.513.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • Terminal-Bench 2.1 (Vals)

    GPT-5.576.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • Claw-Eval

    GPT-5.5—
    Qwen3.5 397B56.8%
    Source

    Not directly comparable

  • QwenClawBench

    GPT-5.5—
    Qwen3.5 397B51.8%
    Source

    Not directly comparable

  • τ³-bench results

    GPT-5.5—
    Qwen3.5 397B68.4%
    Source

    Not directly comparable

  • VITA-Bench

    GPT-5.5—
    Qwen3.5 397B43.7%
    Source

    Not directly comparable

  • DeepPlanning

    GPT-5.5—
    Qwen3.5 397B37.6%
    Source

    Not directly comparable

  • MCP-Tasks

    GPT-5.5—
    Qwen3.5 397B74.2%
    Source

    Not directly comparable

  • WideResearch

    GPT-5.5—
    Qwen3.5 397B74.0%
    Source

    Not directly comparable

Coding

  • SWE-bench Pro

    GPT-5.558.6%
    Source
    Qwen3.5 397B50.9%
    Source

    GPT-5.5 leads this result

  • Terminal-Bench 2.0

    GPT-5.582.0%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • Vibe Code Bench

    GPT-5.569.85%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • React Native Evals

    GPT-5.584.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • cursorBench31

    GPT-5.559.2%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • cursorBench32

    GPT-5.558.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • FrontierCode 1.1 Main

    GPT-5.543.0%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • LiveCodeBench (Vals)

    GPT-5.585.3%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • SWE-bench (Vals)

    GPT-5.582.6%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • SWE-bench Verified

    GPT-5.5—
    Qwen3.5 397B76.2%
    Source

    Not directly comparable

  • LiveCodeBench v6

    GPT-5.5—
    Qwen3.5 397B83.6%
    Source

    Not directly comparable

Reasoning

  • MRCR v2 64K-128K

    GPT-5.583.1%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • MRCR v2 128K-256K

    GPT-5.587.5%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • ARC-AGI-2

    GPT-5.585%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • ARC-AGI-3

    GPT-5.50.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • LongBench v2

    GPT-5.5—
    Qwen3.5 397B63.2%
    Source

    Not directly comparable

  • AI-Needle

    GPT-5.5—
    Qwen3.5 397B68.7%
    Source

    Not directly comparable

Knowledge

  • GPQA

    GPT-5.593.6%
    Source
    Qwen3.5 397B88.4%
    Source

    GPT-5.5 leads this result

  • GPQA-D

    GPT-5.593.6%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • HLE

    GPT-5.552.2%
    Source
    Qwen3.5 397B28.7%
    Source

    GPT-5.5 leads this result

  • HLE w/o tools

    GPT-5.541.4%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • GPQA Diamond (Vals)

    GPT-5.593.2%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • MMLU-Pro (Vals)

    GPT-5.588.1%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • SuperGPQA

    GPT-5.5—
    Qwen3.5 397B70.4%
    Source

    Not directly comparable

  • MMLU-Pro

    GPT-5.5—
    Qwen3.5 397B87.8%
    Source

    Not directly comparable

  • MMLU-Redux

    GPT-5.5—
    Qwen3.5 397B94.9%
    Source

    Not directly comparable

  • C-Eval

    GPT-5.5—
    Qwen3.5 397B93%
    Source

    Not directly comparable

Math

  • FrontierMath (legacy)

    GPT-5.551.7%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    GPT-5.551.700%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • FrontierMath v2 (Tier 4)

    GPT-5.535.400%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • AIME26

    GPT-5.5—
    Qwen3.5 397B93.3%
    Source

    Not directly comparable

  • HMMT Feb 2025

    GPT-5.5—
    Qwen3.5 397B94.8%
    Source

    Not directly comparable

  • HMMT Nov 2025

    GPT-5.5—
    Qwen3.5 397B92.7%
    Source

    Not directly comparable

  • HMMT Feb 2026

    GPT-5.5—
    Qwen3.5 397B87.9%
    Source

    Not directly comparable

  • MMAnswerBench

    GPT-5.5—
    Qwen3.5 397B80.9%
    Source

    Not directly comparable

Multilingual

  • MMLU-ProX

    GPT-5.5—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • NOVA-63

    GPT-5.5—
    Qwen3.5 397B59.1%
    Source

    Not directly comparable

Multimodal

  • MMMU-Pro

    GPT-5.581.2%
    Source
    Qwen3.5 397B79%
    Source

    GPT-5.5 leads this result

  • MMMU-Pro w/ Python

    GPT-5.583.2%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • OfficeQA Pro

    GPT-5.554.1%
    Source
    Qwen3.5 397B—

    Not directly comparable

  • MathVision

    GPT-5.5—
    Qwen3.5 397B88.6%
    Source

    Not directly comparable

  • CharXiv

    GPT-5.5—
    Qwen3.5 397B80.8%
    Source

    Not directly comparable

  • VideoMMMU

    GPT-5.5—
    Qwen3.5 397B84.7%
    Source

    Not directly comparable

  • ScreenSpot Pro

    GPT-5.5—
    Qwen3.5 397B65.6%
    Source

    Not directly comparable

  • V*

    GPT-5.5—
    Qwen3.5 397B95.8%
    Source

    Not directly comparable

Instruction following

  • IFEval

    GPT-5.5—
    Qwen3.5 397B92.6%
    Source

    Not directly comparable

Frequently asked questions

Which is better, GPT-5.5 or Qwen3.5 397B?

GPT-5.5 has the higher public score, 73.27 versus 56.44, and the 90% score intervals do not overlap. That is the clearest overall quality signal in the current public data.

Which is better for coding, GPT-5.5 or Qwen3.5 397B?

GPT-5.5 scores higher for coding on the public lane, 67.7 to 52.4. Qwen3.5 397B is scored on Estimated evidence for coding, so the reading is directional rather than like-for-like. BenchLM does not name a winner for coding from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which is better for agentic tasks, GPT-5.5 or Qwen3.5 397B?

GPT-5.5 scores higher for agentic tasks on the public lane, 63.9 to 48.2. Qwen3.5 397B is scored on Estimated evidence for agentic tasks, so the reading is directional rather than like-for-like. BenchLM does not name a winner for agentic tasks from a directional reading; read the shared benchmark rows directly and test the models on the same task set.

Which costs less, GPT-5.5 or Qwen3.5 397B?

For the stated presets, chat costs $0.02 on GPT-5.5 and $0.0024 on Qwen3.5 397B; repository review costs $0.34 and $0.0408; the cache-heavy agent loop costs $0.5 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, GPT-5.5 or Qwen3.5 397B?

GPT-5.5 has the larger documented context window: 1M, compared with 128K.

Related comparisons

Compare API pricing Read the methodology Open the model selector

Last updated September 4, 2026

Watch GPT-5.5 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.