Agentic
Like-for-like- Gemini 3 Flash
- 34.0
- Supported · #135/153
- GLM-5.2
- 58.5
- Supported · #29/153
- Basis
- BenchAlign lane · 4 vs 6 public rows
- Reading
- GLM-5.2 leads
Keep up with the models you depend on. Follow price changes, retirements, and API updates.Follow the models you depend on.
Follow model changesUpdated September 14, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
Decision reading
GLM-5.2 has the higher public score estimate, 68.19 versus 62.28, but the 90% score intervals overlap. Treat that as a lead, not a settled winner.
5 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
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.
Code generation, repair, and software-engineering tasks
GLM-5.2
GLM-5.2 leads on the public coding lane, 61 to 41.6, with Supported evidence for both models, although the 90% intervals overlap.
Confidence: limited
Tool use, computer use, and multi-step task completion
GLM-5.2
GLM-5.2 leads on the public agentic lane, 58.5 to 34, with Supported evidence for both models and non-overlapping 90% intervals.
Confidence: stronger
1K fresh input + 500 output tokens
Gemini 3 Flash
Gemini 3 Flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
200K cached + 20K fresh input + 10K output tokens
Gemini 3 Flash
Gemini 3 Flash has the lower estimated token cost for this stated workload. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Confidence: rate-fallback
50K fresh input + 3K output tokens
Gemini 3 Flash
Gemini 3 Flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Confidence: listed-rates
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documented
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
1 category rests on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
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.
| Category | Gemini 3 Flash | GLM-5.2 | Basis | Reading |
|---|---|---|---|---|
| Agentic | 34.0Supported · #135/153 | 58.5Supported · #29/153 | Like-for-likeBenchAlign lane · 4 vs 6 public rows | GLM-5.2 leads |
| Coding | 41.6Supported · #109/152 | 61.0Supported · #19/152 | Like-for-likeBenchAlign lane · 3 vs 8 public rows | GLM-5.2 leads · intervals overlap |
| Knowledge | 56.9Supported · #45/183 | 60.7Supported · #35/183 | Like-for-likeBenchAlign lane · 2 vs 6 public rows | GLM-5.2 leads · intervals overlap |
| Instruction following | 66.2#68/123 | 89.8#22/123 | Directional onlyProvisional lane · 0 vs 0 weighted rows | Directional only |
| Reasoning | 59.0Unranked · 2 rankable rows | 74.8Unranked · 2 rankable rows | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 50.2Unranked · 2 rankable rows | 80.7Unranked · 4 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | 75.8Unranked · 1 rankable row | Not ranked | 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.
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.
LiveCodeBench (Vals)
Coding
MMLU-Pro (Vals)
Knowledge
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.
1K fresh input + 500 output tokens
Gemini 3 Flash has the lower modeled cost
Costs use the listed standard API rates.
50K fresh input + 3K output tokens
Gemini 3 Flash has the lower modeled cost
Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
Gemini 3 Flash has the lower modeled cost
GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
Maximum documented context; output-token limits may be lower.
Gemini 3 Flash
GLM-5.2
1M
Gemini 3 Flash
gemini-3-flash-preview
Google Gemini API pricingGLM-5.2
Not sourced
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Gemini 3 Flash
$0.05 per 1M cached input tokens
Google Gemini API pricingGLM-5.2
Not published
Gemini 3 Flash
Not sourced
GLM-5.2
Not sourced
Gemini 3 Flash
Not sourced
GLM-5.2
Not sourced
Gemini 3 Flash
Not sourced
GLM-5.2
Not sourced
Gemini 3 Flash
Non-Reasoning
GLM-5.2
Reasoning
Gemini 3 Flash
Proprietary
GLM-5.2
Open Weight
Gemini 3 Flash
Proprietary
GLM-5.2
Open Weight
Gemini 3 Flash
2025-12-01
GLM-5.2
2026-06-16
Run the same representative tasks against both endpoints before changing production traffic.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Claw-Eval
Not directly comparable
Gert Labs
Not directly comparable
JobBench
Not directly comparable
Terminal-Bench 2.1 (Vals)
GLM-5.2 leads this result
Terminal-Bench 3.0
Not directly comparable
Terminal-Bench 2.0
Not directly comparable
MCP Atlas
Not directly comparable
Toolathlon
Not directly comparable
ResearchClawBench
Not directly comparable
Vibe Code Bench
Not directly comparable
LiveCodeBench (Vals)
Gemini 3 Flash leads this result
SWE-bench (Vals)
GLM-5.2 leads this result
SWE-bench Pro
Not directly comparable
NL2Repo
Not directly comparable
Terminal-Bench 2.0
Not directly comparable
ProgramBench
Not directly comparable
cursorBench32
Not directly comparable
OpenHarmony Bench
Not directly comparable
CritPt
Not directly comparable
GPQA Diamond (Vals)
Gemini 3 Flash leads this result
MMLU-Pro (Vals)
Gemini 3 Flash leads this result
GPQA
Not directly comparable
GPQA-D
Not directly comparable
HLE
Not directly comparable
HLE w/o tools
Not directly comparable
FrontierMath v2 (Tiers 1-3)
Not directly comparable
FrontierMath v2 (Tier 4)
Not directly comparable
AIME26
Not directly comparable
HMMT Nov 2025
Not directly comparable
HMMT Feb 2026
Not directly comparable
MMAnswerBench
Not directly comparable
GLM-5.2 has the higher public score estimate, 68.19 versus 62.28, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
GLM-5.2 leads the public coding lane, 61 to 41.6, with Supported evidence for both models, although the 90% intervals overlap.
GLM-5.2 leads the public agentic tasks lane, 58.5 to 34, with Supported evidence for both models and non-overlapping 90% intervals.
For the stated presets, chat costs $0.002 on Gemini 3 Flash and $0.0036 on GLM-5.2; repository review costs $0.034 and $0.0832; the cache-heavy agent loop costs $0.05 and $0.352. GLM-5.2 has no published cached-input rate, so cached tokens use its listed input rate.
Both models list the same context window, 1M.
Last updated September 14, 2026
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.