Agentic
Like-for-like- Hy3 Preview
- 54.4
- MiniMax M2.7
- 57.0
- Weighted basis
- 1 vs 1 rows
- Reading
- MiniMax M2.7 leads
Model comparison
Hy3 Preview vs MiniMax M2.7
Updated August 7, 2026. Public scores include evidence status and uncertainty. They are not guarantees for a specific workload.
MiniMax M2.7 has the higher public score, 63.1 versus 42.8, and the 90% score intervals do not overlap.
3 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.
Agentic work
Tool use, computer use, and multi-step task completion
MiniMax M2.7
MiniMax M2.7 leads on the same 1 weighted benchmark row.
Confidence: limited
Long documents
Prompts that approach the documented context limit
Hy3 Preview
Hy3 Preview has the larger documented context window.
Confidence: documented
Show secondary and unsupported calls
Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
No shared weighted benchmark basis supports a winner.
Confidence: limited
Chat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
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. 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. Hy3 Preview has no comparable published API token rate.
Confidence: rate-fallback
Repository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
3 Hy3 Preview only
3 Shared
15 MiniMax M2.7 only
- Shared results
- 3
- Hy3 Preview only
- 3
- MiniMax M2.7 only
- 15
- Like-for-like categories
- 1 / 8
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
Not comparable- Hy3 Preview
- 74.4
- MiniMax M2.7
- 53.3
- Weighted basis
- 1 vs 2 rows
- Reading
- Not comparable
Reasoning
Not comparable- Hy3 Preview
- Not measured
- MiniMax M2.7
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Knowledge
Not comparable- Hy3 Preview
- 87.2
- MiniMax M2.7
- Not measured
- Weighted basis
- 1 vs 0 rows
- Reading
- Not comparable
Math
Not comparable- Hy3 Preview
- Not measured
- MiniMax M2.7
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multilingual
Not comparable- Hy3 Preview
- Not measured
- MiniMax M2.7
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Multimodal
Not comparable- Hy3 Preview
- Not measured
- MiniMax M2.7
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
Instruction following
Not comparable- Hy3 Preview
- Not measured
- MiniMax M2.7
- Not measured
- Weighted basis
- 0 vs 0 rows
- Reading
- Not comparable
| Category | Hy3 Preview | MiniMax M2.7 | Weighted basis | Reading |
|---|---|---|---|---|
| Agentic | 54.4 | 57.0 | Like-for-like1 vs 1 rows | MiniMax M2.7 leads |
| Coding | 74.4 | 53.3 | Not comparable1 vs 2 rows | Not comparable |
| Reasoning | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Knowledge | 87.2 | Not measured | Not comparable1 vs 0 rows | Not comparable |
| Math | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multilingual | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Multimodal | Not measured | Not measured | Not comparable0 vs 0 rows | Not comparable |
| Instruction following | Not measured | Not measured | Not comparable0 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
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
- Hy3 Preview
- Self-hosted; infrastructure cost varies
- Fits in one request
- MiniMax M2.7
- $0.0009
- Fits in one request
Hy3 Preview has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Hy3 Preview
- Self-hosted; infrastructure cost varies
- Fits in one request
- MiniMax M2.7
- $0.0186
- Fits in one request
Hy3 Preview has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Hy3 Preview
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
- MiniMax M2.7
- $0.078
- Does not fit in one request
- Cached input priced at the published list-input rate
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. Hy3 Preview has no comparable published API token rate.
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationHy3 PreviewMiniMax M2.7
Context window
Maximum documented context; output-token limits may be lower.
Hy3 Preview
256K
MiniMax M2.7
200K
API model ID
Hy3 Preview
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.
Hy3 Preview
No comparable hosted API rate
MiniMax M2.7
Not published
Documented inputs
Hy3 Preview
Not sourced
MiniMax M2.7
Not sourced
Documented outputs
Hy3 Preview
Not sourced
MiniMax M2.7
Not sourced
Provider availability
Hy3 Preview
Not sourced
MiniMax M2.7
Not sourced
Reasoning profile
Hy3 Preview
Reasoning
MiniMax M2.7
Non-Reasoning
Weight access
Hy3 Preview
Open Weight
MiniMax M2.7
Open Weight
License
Hy3 Preview
Open Weight
MiniMax M2.7
Open Weight
Release date
Hy3 Preview
2026-04-23
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, 63.1 versus 42.8, and the 90% score intervals do not overlap.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Hy3 Preview has the larger documented window (256K).
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
Terminal-Bench 2.0
Hy3 Preview54.4%SourceMiniMax M2.757%SourceMiniMax M2.7 leads this result
Gert Labs
Shared sourceHy3 Preview36.91%MiniMax M2.740.40%MiniMax M2.7 leads this result
Toolathlon
Hy3 Preview—MiniMax M2.746.3%SourceNot directly comparable
MLE-Bench Lite
Hy3 Preview—MiniMax M2.766.6%SourceNot directly comparable
MM-ClawBench
Hy3 Preview—MiniMax M2.762.7%SourceNot directly comparable
Claw-Eval
Hy3 Preview—MiniMax M2.748.7%SourceNot directly comparable
Coding
SWE-bench Verified
Hy3 Preview74.4%SourceMiniMax M2.7—Not directly comparable
Terminal-Bench 2.0
Hy3 Preview54.4%SourceMiniMax M2.7—Not directly comparable
SWE-bench Verified*
Hy3 Preview—MiniMax M2.775.4%SourceNot directly comparable
SWE-bench Pro
Hy3 Preview—MiniMax M2.756.2%SourceNot directly comparable
SWE-Rebench
Hy3 Preview—MiniMax M2.751.9%SourceNot directly comparable
SWE Multilingual
Hy3 Preview—MiniMax M2.776.5%SourceNot directly comparable
Multi-SWE Bench
Hy3 Preview—MiniMax M2.752.7%SourceNot directly comparable
VIBE-Pro
Hy3 Preview—MiniMax M2.755.6%SourceNot directly comparable
NL2Repo
Hy3 Preview—MiniMax M2.739.8%SourceNot directly comparable
Vibe Code Bench
Hy3 Preview—MiniMax M2.727.04%SourceNot directly comparable
React Native Evals
Hy3 Preview—MiniMax M2.771.4%SourceNot directly comparable
Knowledge
Math
AIME25 (Arcee)
Hy3 Preview—MiniMax M2.780.0%SourceNot directly comparable
Frequently asked questions
Which is better, Hy3 Preview or MiniMax M2.7?
MiniMax M2.7 has the higher public score, 63.1 versus 42.8, 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, Hy3 Preview or MiniMax M2.7?
The published evidence does not provide a shared weighted coding basis for both models, so BenchLM does not name a coding winner.
Which is better for agentic tasks, Hy3 Preview or MiniMax M2.7?
MiniMax M2.7 leads the like-for-like agentic tasks comparison across 1 shared weighted benchmark row.
Which costs less, Hy3 Preview or MiniMax M2.7?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, Hy3 Preview or MiniMax M2.7?
Hy3 Preview has the larger documented context window: 256K, compared with 200K.
Related comparisons
Last updated August 7, 2026
Watch Hy3 Preview vs MiniMax M2.7
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 2,000+ readers.