SpringPrompt predicted fit · public beta
Highest-throughput AI models for Email Writing
Predicted ability to write accurate, audience-aware, appropriately toned and action-oriented business, sales and support emails.
gpt-oss-120b has the highest source-listed output throughput among email writing models with an active operational reference.
This is a task-specific prediction from several external benchmark signals. It is not a task-success percentage, guarantee, strict league table, or fresh direct model run. Close scores should be read as the same performance band.
Highest output throughput (beta)
Active reference configurations in the same evidence-qualified task cohort are ranked by source-listed output tokens per second. Deprecated references remain visible as legacy data without a throughput position. This is generation throughput after output begins—not end-to-end response speed—and it does not change predicted fit.
Updated 2026-07-16
| Highest output throughput | Model family | Predicted fit | Blended price† | Output throughput† | Evidence coverage | Independent evidence |
|---|---|---|---|---|---|---|
| #1fit #31 | gpt-oss-120b openai Operational reference: gpt-oss-120b (high) ↗Active configuration | 44.6relative task estimate | $0.20per 1M blended | 255.3 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #2fit #32 | gpt-oss-20b openai Operational reference: gpt-oss-20b (high) ↗Active configuration | 41.8relative task estimate | $0.07per 1M blended | 205.4 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #3fit #18 | Grok 4.5 xai Operational reference: Grok 4.5 (high) ↗Active configuration | 56.7relative task estimate | $1.35per 1M blended | 116.3 tok/sexcludes first-token wait | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #4fit #17 | Gemini 3.1 Pro google Operational reference: Gemini 3.1 Pro Preview ↗Active configuration | 56.8relative task estimate | $1.74per 1M blended | 115.9 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #5fit #15 | DeepSeek-V4-Flash deepseek Operational reference: DeepSeek V4 Flash (max) ↗Active configuration | 57.4relative task estimate | $0.06per 1M blended | 108.6 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #6fit #9 | Claude Sonnet 5 anthropic Operational reference: Claude Sonnet 5 (max) ↗Active configuration | 59.6relative task estimate | $1.54per 1M blended | 70.9 tok/sexcludes first-token wait | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #7fit #2 | GPT 5.5 openai Operational reference: GPT-5.5 (xhigh) ↗Active configuration | 62.9relative task estimate | $4.35per 1M blended | 66.3 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #8fit #3 | Claude Fable 5 anthropic Operational reference: Claude Fable 5 (with fallback) ↗Active configuration | 62.8relative task estimate | $7.70per 1M blended | 65.0 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #9fit #11 | Deepseek V4 Pro deepseek Operational reference: DeepSeek V4 Pro (max) ↗Active configuration | 58.9relative task estimate | $0.18per 1M blended | 62.3 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #10fit #22 | Qwen3.5-397B-A17B alibaba Operational reference: Qwen3.5 397B A17B ↗Active configuration | 53.8relative task estimate | $0.90per 1M blended | 59.6 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #11fit #4 | GPT 5.6 Sol openai Operational reference: GPT-5.6 Sol (max) ↗Active configuration | 62.7relative task estimate | $4.35per 1M blended | 55.9 tok/sexcludes first-token wait | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #12fit #1 | Claude Opus 4.8 anthropic Operational reference: Claude Opus 4.8 (max) ↗Active configuration | 62.9relative task estimate | $3.85per 1M blended | 54.8 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #13fit #5 | Claude Opus 4.7 anthropic Operational reference: Claude Opus 4.7 (max) ↗Active configuration | 62.6relative task estimate | $3.85per 1M blended | 48.3 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| #14fit #14 | Kimi K2.6 moonshotai Operational reference: Kimi K2.6 ↗Active configuration | 58.3relative task estimate | $0.70per 1M blended | 44.4 tok/sexcludes first-token wait | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #6legacy · unranked | GPT 5.4 openai Legacy operational reference: GPT-5.4 (xhigh) ↗Deprecated · legacy data | 61.9relative task estimate | $2.17legacy reference · excluded from filters | 150.3 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #7legacy · unranked | Claude Opus 4.6 anthropic Legacy operational reference: Claude Opus 4.6 (max) ↗Deprecated · legacy data | 61.5relative task estimate | $3.85legacy reference · excluded from filters | 39.0 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #8legacy · unranked | Claude Sonnet 4.6 anthropic Legacy operational reference: Claude Sonnet 4.6 (max) ↗Deprecated · legacy data | 61.1relative task estimate | $2.31legacy reference · excluded from filters | 46.9 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #10legacy · unranked | GPT 5.2 openai Legacy operational reference: GPT-5.2 (xhigh) ↗Deprecated · legacy data | 59.1relative task estimate | $1.87legacy reference · excluded from filters | 65.6 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #12legacy · unranked | GPT 5.1 openai Legacy operational reference: GPT-5.1 (high) ↗Deprecated · legacy data | 58.4relative task estimate | $1.34legacy reference · excluded from filters | 105.5 tok/slegacy reference · excluded from filters | 59% | 3 signal groups · 3 lineages EQ-Bench 3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #13no operational data | Gemini 3 Pro google | 58.3relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #16legacy · unranked | GPT 5 openai Legacy operational reference: GPT-5 (high) ↗Deprecated · legacy data | 56.8relative task estimate | $1.34legacy reference · excluded from filters | 99.0 tok/slegacy reference · excluded from filters | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #19no operational data | Deepseek V3.2 deepseek | 55.8relative task estimate | — | — | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #20legacy · unranked | Kimi K2.5 moonshotai Legacy operational reference: Kimi K2.5 ↗Deprecated · legacy data | 55.4relative task estimate | $0.49legacy reference · excluded from filters | 49.8 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #21no operational data | Deepseek V3.1 deepseek | 54.8relative task estimate | — | — | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #23legacy · unranked | GPT 4.1 openai Legacy operational reference: GPT-4.1 ↗Deprecated · legacy data | 52.4relative task estimate | $1.55legacy reference · excluded from filters | 96.0 tok/slegacy reference · excluded from filters | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #24no operational data | DeepSeek-R1-0528 deepseek | 51.8relative task estimate | — | — | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #25no operational data | DeepSeek-R1 deepseek | 51.1relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #26no operational data | gemma-3-12b-it google | 49.3relative task estimate | — | — | 55% | 3 signal groups · 3 lineages EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #27no operational data | DeepSeek-V3-0324 deepseek | 49.3relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #28no operational data | gemma-3-27b-it google | 48.9relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #29no operational data | qwq-32b alibaba | 47.1relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #30no operational data | gemma-3-4b-it google | 45.2relative task estimate | — | — | 77% | 3 signal groups · 4 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
| —fit #33no operational data | gemma-2-9b-it google | 41.5relative task estimate | — | — | 60% | 3 signal groups · 3 lineages EQ-Bench 3, EQ-Bench Creative Writing v3, LMArena Leaderboard Dataset, UGI Leaderboard |
The provisional fit order is useful for broad comparison, but small numerical differences are not evidence of precise gaps. Predicted-fit scores are meaningful only within this task page. †Operational values use the exact Artificial Analysis reference configuration linked on each row. Operational coverage: 14 active, 8 legacy, and 11 missing out of 33 model families. Only active references receive cost or throughput positions; active price coverage is 14/33 and active throughput coverage is 14/33 in the snapshot dated 2026-07-16. Read the complete methodology →
Publication threshold
A model appears in this beta only after meeting the category's shared evidence floor. For this release that means at least 55% signal coverage, 3 independent signal groups, and 3 evidence lineages. Lower-coverage models are withheld rather than assigned misleading positions.
Illustrative saved outputs
These existing examples show the task shape only. Their earlier judge scores do not contribute to the predicted-fit beta.
Saved example AGoogle: Gemini 3.1 Flash Lite Preview
Subject: Update: Beta invite analytics dashboard release Hi team, The beta invite analytics dashboard did not launch on Tuesday as scheduled. **Cause:** During the event backfill process, we discovered duplicate invite events originating from the legacy importer. This prevented QA from accurately verifying conversion metrics. **Updated Plan:** * **Data Fix:** Lena is resolving the duplication issue by Thursday at 12:00. * **QA:** Omar will perform validation testing Thursday afternoon. * **Release:** Pending successful verification, we will release the dashboard on Friday at 15:00. I am managing stakeholder updates and will notify you immediately if any further blockers arise. Thank you for your patience while we ensure the data accuracy of this release. Best, [Your Name]
Saved example BGoogle: Gemini 3.1 Pro Preview
Subject: Resolving your recurring billing issue Dear [Customer Name], I completely understand your frustration, and you have every right to be upset. Dealing with a single billing error is inconvenient, but having it happen repeatedly is unacceptable. I sincerely apologize for the ongoing hassle and the time you have had to spend addressing this with us. We take full accountability for this recurring issue. To ensure this is permanently resolved, I am taking the following actions immediately: * **Escalation:** I have escalated your account to our billing team to identify the exact root cause of these repeated errors. * **Account Audit:** I am personally reviewing your billing history to ensure all discrepancies are identified and corrected. I will follow up with you directly by [Time/Date] with a full update on what caused this, how we have fixed it, and the next steps regarding your account balance. Thank you for your patience while I work to make this right. Sincerely, [Your Name] [Your Title] [Your Contact Information]
Saved example CGoogle: Gemini 3.1 Pro Preview
**Subject:** Elevate your strategy with [Product Name] **Hi [First Name],** Achieving your core targets requires the right tools. That’s why we developed [Product Name]—to give you a competitive edge and support your ongoing success. We know that every [business/individual] is unique, which is why [Product Name] is built to adapt to your specific needs. By leveraging [Key Feature], our solution is designed to help you: * **Streamline your daily operations:** Reduce friction in your workflow so you can focus on high-impact tasks. * **Target your key pain points:** Equip yourself with tools built specifically to address [Pain Point]. * **Make informed decisions:** Use our integrated insights to guide your next steps. While individual outcomes depend on various factors and specific use cases, our goal is to provide you with a robust framework to help you work smarter and pursue your goals more effectively. Curious to see how [Product Name] can fit into your strategy? **[Link: Explore [Product Name] Today / Book a Demo]** Best regards, [Your Name] [Your Title] [Company Name]
Frequently asked
What is the best AI model for email writing?
Claude Opus 4.8 currently has the highest predicted-fit beta score. This is a relative estimate from several external benchmark signals, not a task-success percentage or a new direct model run.
Does output throughput mean fastest response?
No. Tokens per second measures generation after output begins and excludes time to first token and reasoning delay. Only active reference configurations receive a throughput position; deprecated references remain visible as legacy data. Throughput contributes no quality points.
How is predicted fit calculated?
SpringPrompt normalizes relevant third-party benchmark signals, combines them using reviewed task-specific weights, and keeps evidence coverage separate from performance. Only models meeting the published coverage and independent-evidence floors appear in the beta order.
How precise is the order?
It is a non-strict provisional order. Small score differences should be read as the same performance band, and scores should not be compared across different task pages.
How this beta is calculated
We normalize relevant public benchmark signals within their source, combine them using a reviewed category-specific weighting, and publish evidence coverage alongside the estimate. Missing signals remain missing. Product-family matching follows reviewed identity rules and does not create links to runnable models unless that identity is independently established.
Definition 2026-07-16-business-candidate-v6 · Release external-predictive-public-beta:fd72878b034342eaddb21f5bd557b0c000c016b7194764c76bc5dbd0b3384bae · sources, weights and scoring safeguards