Interactive leaderboard
Fastest LLM APIs 2026: Low-Latency AI Models Compared
Rank the fastest LLM APIs in 2026 using throughput and TTFT latency signals with pricing context. Shortlist responsive AI models for real-time apps in the US, Canada, and Australia.
Speed-ranked LLMs with API cost on the same canvas in 2026
Speed scores reflect interactive-class behavior: smaller fast tiers vs. heavy flagships, grounded in benchmark metadata and tier cues—not a single vendor’s marketing latency claim. Product teams across the United States, Canada, and Australia use this tab to protect UX on chat surfaces while still eyeballing what responsiveness costs at production token volumes.
Workload & pricing toggles
Workload presets
Same three scenarios as the main AI API calculator: moderate traffic, large RAG-style context, or per-request max tokens with a lower request count.
Include Vision / Image Processing
Off — no image fees in cost estimates for vision-capable models.
Turn On to include image fees.
OffOn
Use Cached Pricing
Enable to get 50% off input tokens where cached rates apply
OffOn
Deep Reasoning / Thinking Mode
Model hidden reasoning / extended thinking charged like output tokens when enabled.
OffOn
Batch Pricing
Enable for 50% off input & output where batch/async pricing applies
OffOn
≈ $100.00/mo
8K1K—1.0M
≈ $100.00/mo
2K100—500K
≈ $200.00 total
5K10—100K
Cached / batch est. monthly values only change after the pipeline sets supports_caching or supports_batch in Supabase. The toggles here narrow the table to models whose catalog or provider typically supports those modes.
Magic quadrant (top 15)
X: est. monthly · Y: Speed · Dot: provider color · Hover for rank, model & detailsFull leaderboard
Showing 48 of 513 models.
| Pick | Model | Est. monthly | ROI score | Coding | Reasoning | Speed | Math | Context | Overall |
|---|---|---|---|---|---|---|---|---|---|
| TypeSafe: Jev Router | VARIABLE | 54 | 40 | 45 | 100 | 40 | 1.0M | 43No public SWE-bench or GPQA scores exist; LMMarketCap reports 40/100 for coding. Mapped logic and math to ~40 as a fast System One router. Speed is 100 due to 250k tokens/s and 70ms latency. | |
| Inception: Mercury 3 | $7.50 | 68 | 88 | 88 | 100 | 85 | 1.0M | 87Cited HumanEval 90.0 (Coder Small), plus GPQA Diamond, IFBench, and SWE-Bench Atlas wins over GPT-6 Luna. Mapped to Coding 88, Logic 88, Instruction 88. Speed set to 100 due to unprecedented 1,068 tokens/sec throughput. | |
| Inception: Mercury 2.5 | $3.10 | 71 | 75 | 75 | 100 | 88 | 260K | 78Based on GPQA (73.6-77%), HumanEval (85%), and IFBench (71.3%), mapped to mid-tier 70-80s. Speed is exceptional at ~1,107 tps (100). As a speed-optimized dLLM, it trades flagship coding for extreme throughput and tunable reasoning. | |
| Inception: Mercury 2.5 Preview | $3.10 | 52 | 45 | 45 | 100 | 45 | 260K | 45Fast-dLLM v2 averages 43.5 across HumanEval, GPQA, and MMLU, mapping to ~45 for coding/logic. As a 1B-scale lightweight model, it prioritizes speed (>1,000 tok/s, scoring 100) over flagship-level reasoning. | |
| Inception: Mercury 2 | $17.50 | 39 | 43 | 44 | 100 | 44 | 128K | 44Fast-dLLM v2 (Mercury 2) averages 43.5 across HumanEval, GPQA, and MMLU. As a 1B-scale lightweight model, coding and logic map to ~43. Speed is 100 due to >1,000 tokens/sec diffusion generation. | |
| Relace: Relace Apply 3 | $46.50 | 47 | 70 | 65 | 100 | 50 | 256K | 63Evidence explicitly states benchmarks are unavailable. As a specialized code-patching model with 10,000 tok/s throughput, speed is scored 100. Coding and logic are inferred moderately (70/60) due to lack of SWE-bench or GPQA data. | |
| Google: Gemini 3.1 Flash Lite | $25.00 | 49 | 25 | 83 | 98 | 66 | 1.0M | 64SWE-bench Verified at 22% maps to 25 coding. GPQA Diamond at 86.9% maps to 87 logic. As a Lite tier, it excels in speed (381 t/s, 98) but trails flagships in coding. | |
| Llama Guard 3 8B | $19.66 | 45 | 55 | 60 | 98 | 45 | 131K | 55Based on Llama 3 8B stats: HumanEval 62.2% (Coding 55), MMLU 68.4% (Logic 60), MATH 30% (Math 45). As an 8B lightweight tier, scores are adjusted down versus flagships. Speed is 98 reflecting 765 tps on Groq. | |
| AionLabs: Aion-3.0-Mini | $42.00 | 56 | 75 | 78 | 98 | 85 | 131K | 79Evidence lists SWE-bench and GPQA without exact scores. As a 32B Mini/distilled R1 tier, coding and logic are inferred (~75). Speed is explicitly 1000 tokens/s, mapping to 98. Multimodal unsupported. | |
| Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | $25.00 | 53 | 65 | 81 | 98 | 56 | 66K | 71GPQA Diamond at 86.9% maps to 87 logic. MMMU Pro 76.8% maps to 77 multimodal. Speed is 363 tok/s (98). As a Lite tier, coding is inferred lower (65) lacking SWE-bench data. | |
| Google: Gemini 3.1 Flash Lite Preview | $25.00 | 52 | 55 | 78 | 98 | 70 | 1.0M | 70GPQA Diamond at 86.9% maps logic to 85. Coding score of 30.1 maps to 55. MMMU Pro at 76.8% sets multimodal to 77. As a Lite tier, speed is exceptionally high (381 tok/s) mapping to 98. | |
| LiquidAI: LFM2-24B-A2B | $2.40 | 60 | 71 | 44 | 97 | 55 | 128K | 54Benchable.ai cites Coding 71%, Reasoning 50%, Instruction 38%, and Speed at 97th percentile. Mapped directly to 0-100 scale. As a lightweight 2B active MoE, it prioritizes speed over flagship-level logic and coding. | |
| LiquidAI: LFM2.5-2.6B | $19.00 | 41 | 25 | 53 | 95 | 60 | 128K | 48GSM8K at 82.41% (Math ~60) and IFEval at 79.56% (Instruction ~75). Coding and Logic are low (~25-30) as vendor advises against agentic coding and knowledge tasks. Speed is excellent (95) for this 2.6B lightweight tier. | |
| Anthropic: Claude 3.5 Haiku | $72.00 | 51 | 75 | 73 | 95 | 75 | 200K | 74SWE-bench Verified 40.6% (Coding ~75), MMLU-Pro 65% (Logic ~70), MATH 69.4% (Math ~75). As a lightweight Haiku tier, it scores lower than flagships on reasoning but achieves exceptional speed. | |
| xAI: Grok 3 Mini | $17.00 | 62 | 85 | 85 | 95 | 85 | 131K | 85Web digest notes Grok 3 Mini outperforms Grok-2 mini (HumanEval >87.2%, MATH >70.2%). Mapped to ~85 for coding/math. As a 'Mini' tier with native reasoning, it prioritizes speed (100 tok/s, mapped to 95) over flagship-level logic. | |
| Google: Gemini 3.5 Flash Lite | $37.00 | 53 | 70 | 73 | 95 | 75 | 1.0M | 73Lacking exact 3.5 Flash-Lite scores, inferred from Gemini 3 Flash (SWE-bench Verified 78.0%, GPQA 90.4%). Mapped to 0-100 scale with strict penalties for the 'Lite' tier, reducing coding and logic while maximizing speed. | |
| ScaleDown: Extract | $1.60 | 63 | 40 | 63 | 95 | 40 | 1.0M | 51No specific SWE-bench or GPQA scores cited for ScaleDown Extract. Inferred coding (40) and logic (50) based on its small, task-specific extraction tier. Speed (95) reflects '20X faster' claims. | |
| Google: Gemini Flash Latest | $67.50 | 49 | 70 | 65 | 95 | 75 | 1.0M | 69Evidence cites HumanEval 74.3% (Coding ~70), GPQA 51.0% (Logic ~60), and MMMU 62.3% (Multimodal ~65). As a lightweight Flash tier, scores are adjusted lower than Pro flagships, while Speed is rated high (~95) for its class. | |
| inclusionAI: Ling 3.0 Flash VL | $4.20 | 69 | 85 | 73 | 95 | 90 | 131K | 80SWE-Bench Pro at 56.6% maps to 85 coding. MathArena AIME 93.2 maps to 90 math. Speed is 95 (138.8 t/s). As a Flash-tier MoE, logic (HLE 22.7) and instruction are scaled to 70-75. | |
| Qwen: Qwen3.5-Flash | $5.20 | 74 | 88 | 92 | 95 | 95 | 1.0M | 92SWE-bench Verified at 69.2% maps to 88 coding. GPQA Diamond at 84.2% maps to 92 logic. IFEval 91.9% maps to 92 instruction. As a Flash tier, speed is 95, though its reasoning capabilities rival flagship models. | |
| Google: Gemini 2.5 Flash Lite Preview 09-2025 | $8.00 | 52 | 45 | 67 | 95 | 55 | 1.0M | 58Based on GPQA (65.1-70.9%) and LiveCodeBench (64.1-68.8%), logic and coding map to 68 and 45. As a 'Flash Lite' tier, speed is heavily weighted (95), reflecting its ultra-low latency design over flagship-level reasoning. | |
| Google: Nano Banana 2.1 | $135.00 | 45 | 65 | 60 | 95 | 75 | 66K | 65HumanEval 74.1% (Coding ~65), GPQA 50.3% (Logic ~50), IFEval 74.5% (Instruction ~70). As a Flash-tier model, it prioritizes speed (147 tok/s) over flagship reasoning, reflecting appropriately lower capability scores. | |
| OpenAI: GPT-4o-mini (2024-07-18) | $12.00 | 57 | 70 | 73 | 95 | 75 | 128K | 73Evidence lacks raw benchmarks. Scores inferred cautiously from the 'Mini' lightweight tier profile. Speed is heavily weighted (95), while coding (70) and logic (70) are adjusted downward to reflect its distilled nature compared to flagship models. | |
| OpenAI: GPT-6 Luna Pro | $9.00 | 62 | 85 | 75 | 95 | 75 | 1.1M | 78DeepSWE 66.6% and HumanEval 82.4% map to 85 coding. MMLU-Pro 74.6% maps to 75 logic. Speed of 157 t/s maps to 95. As a lightweight reasoning model, scores reflect its tier despite high SWE-bench performance. | |
| Google: Gemini 3.7 Flash | $67.50 | 49 | 62 | 72 | 95 | 75 | 1.0M | 70SWE-bench Verified 38.0% (mapped to 62) and GPQA 51.0% (mapped to 58). As a Flash-tier model, coding and logic are adjusted lower than heavyweights, while speed (150 tokens/s) is scored highly at 95. | |
| NVIDIA: Nemotron 3.5 Super VL | $11.00 | 69 | 95 | 91 | 95 | 93 | 1.0M | 93SWE-bench Verified at 60.47% maps to 95 coding. GPQA at 79.23% maps to 92 logic. HMMT math is 93.67% (93). Speed is exceptional at 449 tok/s (95). Native reasoning is supported via enable_thinking. | |
| OpenAI: GPT-5.1-Codex-Mini | $30.00 | 58 | 88 | 78 | 95 | 80 | 400K | 81SWE-bench Verified 55.0% (mapped to 88) and GPQA Diamond 52.0% (mapped to 75) show strong capabilities. As a Mini tier, it prioritizes speed (175 tok/s, mapped to 95) while maintaining solid reasoning and coding performance. | |
| xAI: Grok 4 Fast | $13.00 | 40 | 40 | 43 | 95 | 45 | 2.0M | 43Evidence cites a 43.5 average across HumanEval, GPQA, and MMLU for this 1B-scale Fast model. Mapped to ~40-45 for coding and logic. As a lightweight tier, speed is rated very high. | |
| Google: Gemini 3.5 Flash | $150.00 | 56 | 85 | 88 | 95 | 75 | 1.0M | 84SWE-bench Verified 78.0% and GPQA Diamond 90.4% map to 85 and 90. Despite being a lightweight Flash tier (speed 95), explicit evidence dictates high capability scores, though typically lower than Pro. | |
| DeepSeek: DeepSeek V4.1 Flash 1017 | $12.00 | 64 | 79 | 88 | 95 | 80 | 1.0M | 84HumanEval 79.4% and GPQA 90.9% map to 79 and 90. As a Flash tier model, speed is rated high (95), while coding reflects its lightweight nature compared to Pro models. | |
| inclusionAI: Ling 3.1 Flash | $5.25 | 70 | 85 | 80 | 95 | 95 | 262K | 85SWE-bench Pro (56.6%) maps to 85 coding. AIME (93.2) maps to 95 math. As a Flash-tier model, speed is heavily weighted (95), while logic and instruction are inferred lower than flagship counterparts. | |
| MiniMax: MiniMax M3.1 Flash | $12.00 | 57 | 74 | 72 | 95 | 70 | 1.0M | 72DeepSWE v1.1 at 73.7% maps to 74 Coding. GPQA Diamond at 81.3% maps to 81 Logic. As a Flash-tier model, it achieves 200+ TPS (95 Speed) but trades off absolute capability versus heavyweight flagships. | |
| OpenAI: GPT-5.1 Chat | $150.00 | 53 | 85 | 78 | 95 | 80 | 128K | 80HumanEval 91% maps to 85 coding. GPQA 53.6% maps to 75 logic. MMMU 69.1% maps to 75 multimodal. As a lightweight 'Instant' tier model, speed is rated 95, with capabilities adjusted below flagship levels. | |
| Mistral: Ministral 3 8B 2512 | $7.50 | 51 | 45 | 58 | 95 | 65 | 262K | 56GPQA 47.1 and MMLU 76.1% map to Logic 55. LiveCodeBench 30.3 maps to Coding 45. MATH 62.6% maps to Math 65. As an 8B lightweight tier, it scores lower on reasoning but achieves 161 tok/s (Speed 95). | |
| Baidu: Qianfan-OCR-Fast | $55.30 | 36 | 40 | 50 | 95 | 40 | 66K | 45No benchmarks provided. Inferred scores based on 'Fast' tier and OCR specialization. Speed is exceptional (claimed 1M tokens/s). Multimodal scored high for OCR focus; coding and logic scored lower as a specialized lightweight model. | |
| Qwen: Qwen3.6 Flash | $18.75 | 59 | 85 | 78 | 95 | 80 | 1.0M | 80SWE-bench Verified 73.4 (via 35B-A3B base) maps to 85 coding. Flash tier yields 95 speed (119 tok/s). Logic/Math inferred ~75-80 due to missing GPQA/MMLU scores. Native reasoning supported. | |
| Relace: Relace Search | $70.00 | 40 | 50 | 60 | 95 | 40 | 256K | 53No standard benchmarks (SWE-bench, GPQA) provided. Scores estimated for a specialized codebase search subagent. Speed rated 95 based on claims of 10,000 tokens/sec. Native reasoning confirmed via explicit mention of reasoning tokens. | |
| Anthropic: Claude 3 Haiku | $22.50 | 52 | 65 | 73 | 95 | 65 | 200K | 69Claude 3 Haiku (Lightweight tier) scores MMLU 76.7%, HumanEval 75.9%, GSM8K 88.9%, and MMMU 50.2%. Mapped coding to 65 and logic to 70, reflecting its distilled nature compared to flagship models. Speed is rated 95. | |
| Anthropic: Claude Haiku 5.5 | $9.00 | 56 | 68 | 68 | 95 | 70 | 1.0M | 68SWE-bench Verified 40.6% (Coding 68), GPQA 41.6% (Logic 50). Mapped to 0-100 scale. As a lightweight tier model, scores reflect lower reasoning/coding capability versus flagships, while speed (104 tokens/s) is heavily weighted higher. | |
| Meta: Llama 3.2 1B Instruct | $3.09 | 45 | 25 | 35 | 95 | 30 | 60K | 31MMLU 49.3%, GSM8K 44.4%, MATH 30.6%. As a 1B lightweight model, Logic (35) and Math (30) reflect sub-50% benchmarks. Coding (25) based on 0.6 index. Speed (95) is maximized for this ultra-small tier. | |
| Google: Nano Banana (Gemini 2.5 Flash Image) | $37.00 | 36 | 40 | 45 | 95 | 40 | 33K | 43Evidence explicitly states 'Benchmark not available' for MMLU/MMMU/GSM8K. Inferred Flash-tier baseline scores (40-50). Speed scored 95 due to 172 tok/s throughput. Multimodal inferred at 70 for a lightweight image model. | |
| OpenAI: gpt-oss-120b | $12.00 | 59 | 70 | 80 | 95 | 75 | 131K | 76HumanEval 71% maps to 70 coding. MMLU 66-90% maps to 80 logic. GSM8K 75% maps to 75 math. 500 tok/s throughput maps to 95 speed. Native reasoning supported via OpenRouter reasoning parameter. | |
| Qwen: Qwen3.8 Omni Flash | $10.70 | 66 | 85 | 88 | 95 | 87 | 1.0M | 87SWE-bench Pro 63.3% maps to 85 coding. GPQA Diamond 91.0% maps to 91 logic. As a Flash tier model, speed is rated high (95), while coding/logic reflect its lightweight nature despite strong reasoning capabilities. | |
| Z.ai: GLM 5.3 FlashX | $27.30 | 52 | 60 | 73 | 95 | 75 | 1.0M | 70DeepSWE 63.4% and MMLU 88.1% map to 60 Coding and 75 Logic. As a Flash tier model, capabilities are adjusted lower than the GLM-5 flagship. Speed is rated 95 based on 200 tokens/s throughput. | |
| Google: Gemini 3 Flash Preview | $50.00 | 60 | 88 | 89 | 95 | 85 | 1.0M | 88GPQA Diamond 90.4% maps to Logic 92; SWE-bench 78% maps to Coding 88. As a Flash-tier model, Speed is rated very high (95). Multimodal inferred at 80 due to extensive video/audio/image support. | |
| OpenAI: GPT-5.4 Mini | $75.00 | 54 | 70 | 80 | 95 | 80 | 400K | 78No exact benchmarks for GPT-5.4 Mini in evidence. Inferred coding (70) and logic (75) based on its 'Mini' tier status and reasoning capabilities. Speed (95) reflects high-throughput optimization. Vision price estimated from $0.75/1M input cost. | |
| OpenAI: GPT-6 Luna | $9.00 | 61 | 85 | 75 | 95 | 70 | 1.1M | 76DeepSWE 1.1 at 66.6% and HumanEval at 82.4% map to 85 coding. MMLU-Pro at 74.6% maps to 75 logic. As a lightweight 'Luna' tier, scores reflect its cost-efficiency focus despite high reasoning effort. | |
| Qwen: Qwen3 Coder Flash | $17.55 | 53 | 70 | 68 | 95 | 65 | 1.0M | 68Evidence lacks exact benchmark numbers but notes Qwen3 Coder Flash is a speed-optimized, lightweight tier. Scores inferred cautiously for a Flash model, prioritizing speed (95) over coding/logic compared to the flagship Coder Plus. |
Need a shareable artifact?
Get a print-ready PDF of your results and a CSV spreadsheet. Tap the button, then enter your work email. We use it to build your files and start the download—and to email you a copy if the site owner enabled that.
AI ROI Leaderboard & Discovery by LeadsCalc
Detailed analysis
PDF Breakdown
Receive a comprehensive native vector PDF of this leaderboard: your workload, filters, top rankings, and a table snapshot (sorted: Speed).
Instant setupNo CC required
By submitting, you agree to our Privacy Policy and Terms.
Agency accelerator
Whitelabel Speed Leaderboard
for your site
Embed the interactive speed view on your own domain — whitelabel branding, lead capture, and the same workload sliders your prospects already use on LeadsCalc.
1-Click CRM sync
Custom branding
Branded reports
Lead analytics