Agentic
Directional only- GPT-5.5 Pro
- 60.6
- Estimated · #24/151
- Qwen3.5 397B
- 48.2
- Estimated · #77/151
- Basis
- BenchAlign lane · 1 vs 13 public rows
- Reading
- Directional only
GPT-5.5 Pro vs Qwen3.5 397B
Updated September 4, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GPT-5.5 Pro has the higher public score estimate, 63.07 versus 56.44, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
2 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
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 Pro
GPT-5.5 Pro 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
GPT-5.5 Pro is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
GPT-5.5 Pro and Qwen3.5 397B are 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. GPT-5.5 Pro has no published cached-input rate, so cached tokens use its listed input rate. 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.
4 GPT-5.5 Pro only
2 Shared
36 Qwen3.5 397B only
- Shared results
- 2
- GPT-5.5 Pro only
- 4
- Qwen3.5 397B only
- 36
- Like-for-like categories
- 0 / 8
2 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.
Knowledge
Directional only- GPT-5.5 Pro
- 61.1
- Estimated · #36/181
- Qwen3.5 397B
- 51.6
- Estimated · #80/181
- Basis
- BenchAlign lane · 2 vs 6 public rows
- Reading
- Directional only
Coding
Not comparable- GPT-5.5 Pro
- Not ranked
- Qwen3.5 397B
- 52.4
- Estimated · #56/183
- Basis
- BenchAlign lane · 0 vs 3 public rows
- Reading
- Not comparable
Reasoning
Not comparable- GPT-5.5 Pro
- Not ranked
- Qwen3.5 397B
- 59.8
- Unranked · 2 rankable rows
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Math
Not comparable- GPT-5.5 Pro
- 70.2
- 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 Pro
- Not ranked
- Qwen3.5 397B
- 69.7
- #5/12
- Basis
- Provisional lane · 0 vs 1 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- GPT-5.5 Pro
- Not ranked
- Qwen3.5 397B
- 63.0
- #27/48
- Basis
- Provisional lane · 0 vs 2 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- GPT-5.5 Pro
- Not ranked
- Qwen3.5 397B
- 0.0
- #120/120
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | GPT-5.5 Pro | Qwen3.5 397B | Basis | Reading |
|---|---|---|---|---|
| Agentic | 60.6Estimated · #24/151 | 48.2Estimated · #77/151 | Directional onlyBenchAlign lane · 1 vs 13 public rows | Directional only |
| Knowledge | 61.1Estimated · #36/181 | 51.6Estimated · #80/181 | Directional onlyBenchAlign lane · 2 vs 6 public rows | Directional only |
| Coding | Not ranked | 52.4Estimated · #56/183 | Not comparableBenchAlign lane · 0 vs 3 public rows | Not comparable |
| Reasoning | Not ranked | 59.8Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Math | 70.2Unranked · 3 rankable rows | 74.2Unranked · 5 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | 69.7#5/12 | Not comparableProvisional lane · 0 vs 1 weighted rows | Not comparable |
| Multimodal | Not ranked | 63.0#27/48 | Not comparableProvisional lane · 0 vs 2 weighted rows | Not comparable |
| Instruction following | Not ranked | 0.0#120/120 | Not comparableProvisional lane · 0 vs 0 weighted rows | 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.
HLE
Knowledge
BrowseComp
Agentic
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 Pro
- $0.12
- 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 Pro
- $2.04
- 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 Pro
- $8.40
- Fits in one request
- Cached input priced at the published list-input rate
- 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. GPT-5.5 Pro has no published cached-input rate, so cached tokens use its listed input rate. 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.
SpecificationGPT-5.5 ProQwen3.5 397B
Context window
Maximum documented context; output-token limits may be lower.
GPT-5.5 Pro
Qwen3.5 397B
128K
API model ID
GPT-5.5 Pro
gpt-5.5-pro
OpenAI GPT-5.5 Pro model documentationQwen3.5 397B
Not sourced
Cached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-5.5 Pro
Not published
OpenAI pricingQwen3.5 397B
Not published
Documented inputs
GPT-5.5 Pro
text, image
OpenAI model catalogQwen3.5 397B
Not sourced
Documented outputs
GPT-5.5 Pro
Qwen3.5 397B
Not sourced
Provider availability
GPT-5.5 Pro
Generally Available · OpenAI Responses API
OpenAI model catalogQwen3.5 397B
Not sourced
Reasoning profile
GPT-5.5 Pro
Reasoning
Qwen3.5 397B
Non-Reasoning
Weight access
GPT-5.5 Pro
Proprietary
Qwen3.5 397B
Open Weight
License
GPT-5.5 Pro
Proprietary
Qwen3.5 397B
Open Weight
Release date
GPT-5.5 Pro
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 Pro has the higher public score estimate, 63.07 versus 56.44, but the 90% score intervals overlap.
- Workload cost
- Repository review: $2.04 vs $0.0408. Cache-heavy agent loop: $8.40 vs $0.168.
- Context tradeoff
- GPT-5.5 Pro has the larger documented window (1.05M).
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 evidence42 rows
Agentic
BrowseComp
GPT-5.5 Pro90.1%SourceQwen3.5 397B62%SourceGPT-5.5 Pro leads this result
Terminal-Bench 2.0
GPT-5.5 Pro—Qwen3.5 397B52.5%SourceNot directly comparable
Claw-Eval
GPT-5.5 Pro—Qwen3.5 397B56.8%SourceNot directly comparable
QwenClawBench
GPT-5.5 Pro—Qwen3.5 397B51.8%SourceNot directly comparable
τ³-bench results
GPT-5.5 Pro—Qwen3.5 397B68.4%SourceNot directly comparable
VITA-Bench
GPT-5.5 Pro—Qwen3.5 397B43.7%SourceNot directly comparable
DeepPlanning
GPT-5.5 Pro—Qwen3.5 397B37.6%SourceNot directly comparable
Toolathlon
GPT-5.5 Pro—Qwen3.5 397B36.3%SourceNot directly comparable
MCP Atlas
GPT-5.5 Pro—Qwen3.5 397B46.1%SourceNot directly comparable
MCP-Tasks
GPT-5.5 Pro—Qwen3.5 397B74.2%SourceNot directly comparable
WideResearch
GPT-5.5 Pro—Qwen3.5 397B74.0%SourceNot directly comparable
Gert Labs
GPT-5.5 Pro—Qwen3.5 397B46.76%SourceNot directly comparable
ResearchClawBench
GPT-5.5 Pro—Qwen3.5 397B14.2%SourceNot directly comparable
Coding
Reasoning
Knowledge
HLE
GPT-5.5 Pro57.2%SourceQwen3.5 397B28.7%SourceGPT-5.5 Pro leads this result
HLE w/o tools
GPT-5.5 Pro43.1%SourceQwen3.5 397B—Not directly comparable
GPQA
GPT-5.5 Pro—Qwen3.5 397B88.4%SourceNot directly comparable
SuperGPQA
GPT-5.5 Pro—Qwen3.5 397B70.4%SourceNot directly comparable
MMLU-Pro
GPT-5.5 Pro—Qwen3.5 397B87.8%SourceNot directly comparable
MMLU-Redux
GPT-5.5 Pro—Qwen3.5 397B94.9%SourceNot directly comparable
C-Eval
GPT-5.5 Pro—Qwen3.5 397B93%SourceNot directly comparable
Math
FrontierMath (legacy)
GPT-5.5 Pro52.4%SourceQwen3.5 397B—Not directly comparable
FrontierMath v2 (Tiers 1-3)
GPT-5.5 Pro51.000%SourceQwen3.5 397B—Not directly comparable
FrontierMath v2 (Tier 4)
GPT-5.5 Pro39.600%SourceQwen3.5 397B—Not directly comparable
AIME26
GPT-5.5 Pro—Qwen3.5 397B93.3%SourceNot directly comparable
HMMT Feb 2025
GPT-5.5 Pro—Qwen3.5 397B94.8%SourceNot directly comparable
HMMT Nov 2025
GPT-5.5 Pro—Qwen3.5 397B92.7%SourceNot directly comparable
HMMT Feb 2026
GPT-5.5 Pro—Qwen3.5 397B87.9%SourceNot directly comparable
MMAnswerBench
GPT-5.5 Pro—Qwen3.5 397B80.9%SourceNot directly comparable
Multilingual
Multimodal
MMMU-Pro
GPT-5.5 Pro—Qwen3.5 397B79%SourceNot directly comparable
MathVision
GPT-5.5 Pro—Qwen3.5 397B88.6%SourceNot directly comparable
CharXiv
GPT-5.5 Pro—Qwen3.5 397B80.8%SourceNot directly comparable
VideoMMMU
GPT-5.5 Pro—Qwen3.5 397B84.7%SourceNot directly comparable
ScreenSpot Pro
GPT-5.5 Pro—Qwen3.5 397B65.6%SourceNot directly comparable
V*
GPT-5.5 Pro—Qwen3.5 397B95.8%SourceNot directly comparable
Instruction following
IFEval
GPT-5.5 Pro—Qwen3.5 397B92.6%SourceNot directly comparable
Frequently asked questions
Which is better, GPT-5.5 Pro or Qwen3.5 397B?
GPT-5.5 Pro has the higher public score estimate, 63.07 versus 56.44, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GPT-5.5 Pro or Qwen3.5 397B?
GPT-5.5 Pro is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, GPT-5.5 Pro or Qwen3.5 397B?
GPT-5.5 Pro scores higher for agentic tasks on the public lane, 60.6 to 48.2. GPT-5.5 Pro and Qwen3.5 397B are 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 Pro or Qwen3.5 397B?
For the stated presets, chat costs $0.12 on GPT-5.5 Pro and $0.0024 on Qwen3.5 397B; repository review costs $2.04 and $0.0408; the cache-heavy agent loop costs $8.40 and $0.168. Qwen3.5 397B does not fit this workload in one request. GPT-5.5 Pro has no published cached-input rate, so cached tokens use its listed input rate. 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 Pro or Qwen3.5 397B?
GPT-5.5 Pro has the larger documented context window: 1.05M, compared with 128K.
Related comparisons
Last updated September 4, 2026
Watch GPT-5.5 Pro vs Qwen3.5 397B
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.