Coding
Like-for-like- GLM-5
- 66.3
- GPT-5.3 Codex
- 67.2
- Weighted basis
- 3 vs 3 rows
- Reading
- GPT-5.3 Codex leads
GLM-5 vs GPT-5.3 Codex
Updated September 3, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GLM-5 has the higher public score estimate, 65.68 versus 65.52, 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.
Coding work
Code generation, repair, and software-engineering tasks
GPT-5.3 Codex
GPT-5.3 Codex leads on the same 3 weighted benchmark rows.
Confidence: stronger
Long documents
Prompts that approach the documented context limit
GPT-5.3 Codex
GPT-5.3 Codex has the larger documented context window.
Confidence: documented
Chat turn cost
1K fresh input + 500 output tokens
GLM-5
GLM-5 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
GLM-5
GLM-5 has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
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
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. GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate. GPT-5.3 Codex 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.
31 GLM-5 only
5 Shared
3 GPT-5.3 Codex only
- Shared results
- 5
- GLM-5 only
- 31
- GPT-5.3 Codex only
- 3
- Like-for-like categories
- 1 / 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.
Agentic
Directional only- GLM-5
- 56.2
- GPT-5.3 Codex
- 71.4
- Weighted basis
- 1 vs 2 rows
- Reading
- Directional only
Reasoning
Not comparable- GLM-5
- 60.8
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Knowledge
Not comparable- GLM-5
- 66.4
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 4 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- GLM-5
- 56.3
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 4 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- GLM-5
- 83.1
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- GLM-5
- Not measured
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- GLM-5
- 92.6
- GPT-5.3 Codex
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
| Category | GLM-5 | GPT-5.3 Codex | Weighted basis | Reading |
|---|---|---|---|---|
| Coding | 66.3 | 67.2 | Like-for-like3 vs 3 rows | GPT-5.3 Codex leads |
| Agentic | 56.2 | 71.4 | Directional only1 vs 2 rows | Directional only |
| Reasoning | 60.8 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Knowledge | 66.4 | Not measured | Not comparable4 vs 0 rows | Not comparable |
| Math | 56.3 | Not measured | Not comparable4 vs 0 rows | Not comparable |
| Multilingual | 83.1 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Multimodal | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Instruction following | 92.6 | Not measured | Not comparable1 vs 0 rows | 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.
Terminal-Bench 2.0
Agentic
SWE-bench Verified
Coding
SWE-Rebench
Coding
SWE-bench Pro
Coding
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
- GLM-5
- $0.0026
- Fits in one request
- GPT-5.3 Codex
- $0.00875
- Fits in one request
GLM-5 has the lower modeled cost
Costs use the listed standard API rates.
Repository review
50K fresh input + 3K output tokens
- GLM-5
- $0.0596
- Fits in one request
- GPT-5.3 Codex
- $0.1295
- Fits in one request
GLM-5 has the lower modeled cost
Costs use the listed standard API rates.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- GLM-5
- $0.252
- Does not fit in one request
- Cached input priced at the published list-input rate
- GPT-5.3 Codex
- $0.525
- Fits in one request
- Cached input priced at the published list-input rate
GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate. GPT-5.3 Codex 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.
SpecificationGLM-5GPT-5.3 Codex
Context window
Maximum documented context; output-token limits may be lower.
GLM-5
200K
GPT-5.3 Codex
API model ID
GLM-5
Not sourced
GPT-5.3 Codex
gpt-5.3-codex
OpenAI GPT-5.3 Codex model documentationCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GLM-5
Not published
GPT-5.3 Codex
Not published
Documented inputs
GLM-5
Not sourced
GPT-5.3 Codex
text, image
OpenAI model catalogDocumented outputs
GLM-5
Not sourced
GPT-5.3 Codex
Provider availability
GLM-5
Not sourced
GPT-5.3 Codex
Generally Available · OpenAI Responses API
OpenAI model catalogReasoning profile
GLM-5
Non-Reasoning
GPT-5.3 Codex
Reasoning
Weight access
GLM-5
Open Weight
GPT-5.3 Codex
Proprietary
License
GLM-5
Open Weight
GPT-5.3 Codex
Proprietary
Release date
GLM-5
2026-03-01
GPT-5.3 Codex
2026-02-05
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
- GLM-5 has the higher public score estimate, 65.68 versus 65.52, but the 90% score intervals overlap.
- Workload cost
- Repository review: $0.0596 vs $0.1295. Cache-heavy agent loop: $0.252 vs $0.525.
- Context tradeoff
- GPT-5.3 Codex has the larger documented window (400K).
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 evidence39 rows
Agentic
Terminal-Bench 2.0
GLM-556.2%SourceGPT-5.3 Codex77.3%SourceGPT-5.3 Codex leads this result
Claw-Eval
GLM-557.7%SourceGPT-5.3 Codex—Not directly comparable
QwenClawBench
GLM-554.1%SourceGPT-5.3 Codex—Not directly comparable
τ³-bench results
GLM-565.6%SourceGPT-5.3 Codex—Not directly comparable
DeepPlanning
GLM-514.6%SourceGPT-5.3 Codex—Not directly comparable
Toolathlon
GLM-538%SourceGPT-5.3 Codex—Not directly comparable
MCP Atlas
GLM-531.1%SourceGPT-5.3 Codex—Not directly comparable
MCP-Tasks
GLM-560.8%SourceGPT-5.3 Codex—Not directly comparable
WideResearch
GLM-569.8%SourceGPT-5.3 Codex—Not directly comparable
CyberGym
GLM-543.2%SourceGPT-5.3 Codex—Not directly comparable
Gert Labs
Shared sourceGLM-550.99%GPT-5.3 Codex57.47%GPT-5.3 Codex leads this result
OSWorld-Verified
GLM-5—GPT-5.3 Codex64.7%SourceNot directly comparable
JobBench
GLM-5—GPT-5.3 Codex33.7%SourceNot directly comparable
Coding
SWE-bench Verified
GLM-577.8%SourceGPT-5.3 Codex85%SourceGPT-5.3 Codex leads this result
SWE-bench Verified*
GLM-572.8%SourceGPT-5.3 Codex—Not directly comparable
SWE-bench Pro
GLM-555.1%SourceGPT-5.3 Codex56.8%SourceGPT-5.3 Codex leads this result
SWE Multilingual
GLM-573.3%SourceGPT-5.3 Codex—Not directly comparable
SWE-Rebench
Shared sourceGLM-562.8%GPT-5.3 Codex58.2%GLM-5 leads this result
React Native Evals
GLM-574.8%SourceGPT-5.3 Codex—Not directly comparable
Vibe Code Bench
GLM-5—GPT-5.3 Codex61.77%SourceNot directly comparable
Reasoning
Knowledge
GPQA
GLM-586%SourceGPT-5.3 Codex—Not directly comparable
GPQA-D
GLM-586.0%SourceGPT-5.3 Codex—Not directly comparable
SuperGPQA
GLM-566.8%SourceGPT-5.3 Codex—Not directly comparable
MMLU-Pro
GLM-585.7%SourceGPT-5.3 Codex—Not directly comparable
MMLU-Pro (Arcee)
GLM-585.8%SourceGPT-5.3 Codex—Not directly comparable
HLE
GLM-550.4%SourceGPT-5.3 Codex—Not directly comparable
Math
AIME26
GLM-595.8%SourceGPT-5.3 Codex—Not directly comparable
AIME25 (Arcee)
GLM-593.3%SourceGPT-5.3 Codex—Not directly comparable
HMMT Feb 2025
GLM-597.5%SourceGPT-5.3 Codex—Not directly comparable
HMMT Nov 2025
GLM-596.9%SourceGPT-5.3 Codex—Not directly comparable
HMMT Feb 2026
GLM-586.4%SourceGPT-5.3 Codex—Not directly comparable
MMAnswerBench
GLM-582.5%SourceGPT-5.3 Codex—Not directly comparable
FrontierMath v2 (Tiers 1-3)
GLM-516.434%SourceGPT-5.3 Codex—Not directly comparable
FrontierMath v2 (Tier 4)
GLM-52.100%SourceGPT-5.3 Codex—Not directly comparable
Multilingual
Instruction following
IFEval
GLM-592.6%SourceGPT-5.3 Codex—Not directly comparable
Frequently asked questions
Which is better, GLM-5 or GPT-5.3 Codex?
GLM-5 has the higher public score estimate, 65.68 versus 65.52, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
Which is better for coding, GLM-5 or GPT-5.3 Codex?
GPT-5.3 Codex leads the like-for-like coding comparison across 3 shared weighted benchmark rows.
Which is better for agentic tasks, GLM-5 or GPT-5.3 Codex?
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, GLM-5 or GPT-5.3 Codex?
For the stated presets, chat costs $0.0026 on GLM-5 and $0.00875 on GPT-5.3 Codex; repository review costs $0.0596 and $0.1295; the cache-heavy agent loop costs $0.252 and $0.525. GLM-5 does not fit this workload in one request. GLM-5 has no published cached-input rate, so cached tokens use its listed input rate. GPT-5.3 Codex has no published cached-input rate, so cached tokens use its listed input rate.
Which has the larger context window, GLM-5 or GPT-5.3 Codex?
GPT-5.3 Codex has the larger documented context window: 400K, compared with 200K.
Related comparisons
Last updated September 3, 2026
Watch GLM-5 vs GPT-5.3 Codex
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.